INTER AMBO MARIA Northern Barbarians from Scandinavia towards the Black Sea

Transcription

1 Vest-Agder County Council V. I. Vernadsky National Taurida University Heritage of Millennia Non-Profitable Foundation for History and Archaeology INTER AMBO MARIA Northern Barbarians from Scandinavia towards the Black Sea Kristiansand Simferopol DOLYA Publishing House 2013

2 ISSN Scholarly editors: Igor Khrapunov and Frans-Arne Stylegar The present volume is number 15 in the series Cultural-Historical Reports (Kulturhistoriske rapporter) published by Vest-Agder County Council Inter Ambo Maria: Northern Barbarians from Scandinavia towards the Black Sea / edited by Igor Khrapunov and Frans-Arne Stylegar. Kristiansand Simferopol: Dolya Publishing House, p. (Cultural-historical reports. 15 / Kulturhistoriske rapporter. 15). Includes bibliographical references. ISSN The conference Inter Ambo Maria: Northern Barbarians from Scandinavia towards the Black Sea was held on October 3 rd 7 th, 2012 in the village of Gaspra (Crimea, Ukraine). It was organised by Vest-Agder County Council (Kristiansand, Norway), Heritage of Millennia Foundation, and V. I. Vernadsky National Taurida University (Simferopol, Ukraine). Researchers of 11 European countries Ukraine, Russia, Poland, Hungary, Slovakia, Czech Republic, Germany, France, Denmark, Sweden, and Norway came to the Crimea to discuss aspects of ethnic and cultural interrelations in the inter ambo maria zone between the Baltic and the Black Seas in the Late Roman and Migration periods. Following the discussion of the presentations, their authors have revised them to be published as papers within this volume. The first conference Inter Ambo Maria took place in Gaspra in October Its proceedings were published in the volume entitled Inter Ambo Maria: Contacts between Scandinavia and the Crimea in Roman Period (Kristiansand; Simferopol, 2011). Book cover concept: Maxim Levada Book Layout Boris Prokopenko Printed in: Format: 60 x84 1 / 8 Font: Times New Roman All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without the prior permission of the publisher. Vest-Agder County Council, 2013 Heritage of Millennia Foundation, 2013 Authors, 2013

3 Editorial The conference Inter Ambo Maria: Northern Barbarians from Scandinavia towards the Black Sea was held on October 3 rd 7 th, 2012 in the village of Gaspra near Yalta in the Crimea, Ukraine. The second in a series of conferences, it turned out to be a major gathering of scholars from different traditions of research, both East and West. The initiative to establish a conference series on the topic of the contacts between the northern Black Sea area and Northern Europe (inter ambo maria) at the end of Antiquity and the Early Medieval period came from Vest-Agder County Council, Norway. The Council s support allowed the Heritage of Millennia Historical and Archaeological Foundation and V. I. Vernadsky National Taurida University to convene researchers of eleven different count ries (Czech Republic, Denmark, France, Germany, Hungary, Norway, Poland, Russia, Slovakia, Sweden, and Ukraine) in the Crimea, where they discussed aspects of ethnic and cultural interrelations in the inter ambo maria zone in the Late Roman and Migration periods. The first conference Inter Ambo Maria: Contacts between Scandinavia and the Crimea in the Roman Period took place in Gaspra in October 2010, and its proceedings were published as a special volume of the same name (Kristiansand; Simferopol, 2011). The 2012 conference included 22 oral presentations, and 16 more participants presented posters. The participants have revised their presentations to be published as papers within this volume. Hopefully, this collection will not disappoint our readers! In 2014, the Crimea will once again see a gathering of archaeologists from different parts of Europe, and thus bring the Baltic and North Seas closer to the Black Sea at least on the mental maps. See you there! Igor Khrapunov Frans-Arne Stylegar Editors Participants of the Inter Ambo Maria conference visiting a fortress of mediaeval Crimean Goths a top Mangup mountain in the Crimea

4 4 Translation and Transliteration Translation and Transliteration The translation of this book follows the pattern used in other volumes published by the Heritage of Millennia Foundation and Vest-Agder County Council, and namely the proceedings of the first Inter Ambo Maria conference and Black Sea Archaeology in Translation series. Our intention was consistency although it is hardly possible to be consistent in all the cases. All the personal, ethnic and geographic names are given in common forms, for example provided by the Encyclopædia Britannica. Because not every name and term used in this volume can be found in the Encyclopædia, it was decided to transliterate them using the method of giving the closest equivalent to Greek, Russian, or Ukrainian pronunciation similar to the patterns of the Chicago style. For example, for Russian alphabet: а a, б b, в v, г g, д d, е ye (open syllable, or after soft and hard signs) or e (close syllable), ж zh, з z, и i, й y, к k, л l, м m, н n, о o, п p, р r, с s, т t, у u, ф f, х kh, ц ts, ч ch, ш sh, щ shch, ъ, ы y, ь, э e, ю yu, я ya. As for the colleagues names, there are few cases when we deviated from this pattern if we definitely know that he or she prefers alternative spelling. And besides, sometimes scholars names are given in one form within the paper and in the other in bibliography. A sad but usual situation with people of East Slavonic origin. That is actually because different publishers use different systems of transliteration. Names of Greek origin are transliterated from Greek (i. e. Tiberios Ioulios Eupatoros), of Latin from Latin (Julius Caesar), except for those having traditional spelling (Diocletian). References to classical authors are given according to the list of abbreviations in the second edition of The Oxford Classical Dictionary, followed by the appropriate book, chapter, paragraph, or line numbers, separated by periods. As for the origin of many place names in Eastern Europe it is not quite clear whether it is Russian or Ukrainian, and the pronunciations in these languages differ, it is decided to use Russian as the basis. A few words (mainly terms) that are absent in English are italicized. English texts of papers by Blankenfeldt, Boye Lund Hansen, Czarnecka, Fabech Näsman, Grane, Istvánovits Kulcsár, Kontny, Prochowicz, Stylegar, Tejral, and Woźniak are provided by authors, paper by Andrzejowski Madyda-Legutko is translated from Polish to English by Magdalena Małek, the paper by Magomedov is translated from Russian to English by Lidia Orlova, paper by Mączyńska is translated from Polish original to Russian by Maxim Levada and from Russian to English by Nikita Khrapunov. Papers by Gavritukhin, Kazanski, Khrapunov, Levada, Lyubichev, Mastykova Dobrovol skaya, Oblomskiy, Radyush, Shabanov, Skvortsov, Terpilovskiy Zharov, Vasil yev, and the Editorial are translated from Russian by Nikita Khrapunov. Russian abstracts to papers by Gavritukhin, Istvánovits Kulcsár, Kazanski, Khrapunov, Levada, Lyubichev, Mastykova Dobrovol skaya, Levada, Magomedov, Oblomskiy, Prochowicz, Radyush, Shabanov, Skvortsov, Terpilovskiy Zharov, and Vasil yev are provided by authors, abstract to paper by Mączyńska is translated from Polish to Russian by Maxim Levada, and abstracts to papers by Andrzejowski Madyda-Legutko, Blankenfeldt, Boye Lund Hansen, Czarnecka, Fabech Näsman, Grane, Istvánovits Kulcsár, Kontny, Stylegar, Tejral, and Woźniak are translated from English to Russian by Nikita Khrapunov. The translator is deeply indebted to Maxim Levada for his varied, however always kind and friendly help with the terms, names, and other specifically-archaeological questions.

5 Abbreviations 5 Abbreviations AAH Acta Archaeologica Academiae Scientiarum Hungaricae AAustr Archaeoloia Austriaca APA Acta Praehistorica et Archaeologica AÚ AV ČR Archeologický Ústav Akademie věd České republiky Brno AÚ SAV Archeologický Ústav Slovenskej Akadémie vied BAR IS British Archaeological Reports. International Series CIRB Корпус боспорских надписей. М.; Л., 1965 IOSPE I 2 Latyshev B. Inscriptiones antiquae orae septentrionales Ponti Euxini grecae et latinae. I. 2nd ed. Saint Petersburg, 1916 Jahrbuch des RGZM Jahrbuch des Römisch-Germanischen Zentralmuseums Mainz JGS Journal of the Glass Studies PLRE Prosopography of the Later Roman Empire. 1. Cambridge, 1971 RIC The Roman Imperial Coinage SPÖG Schriften der königlichen physikalisch-ökonomischer Gesselschaft zu Königsberg TM Travaux et mémoires ZNUJ PA Zeszyty naukowe Uniwersytetu Jagiellońskiego. Prace archeologiczne АВ Археологические вести БИ Боспорские исследования БСб Боспорский сборник ВДИ Вестник древней истории ДБ Древности Боспора ИАК Известия Императорской археологической комиссии КСИА Краткие сообщения Института археологии Л Ленинград М Москва МАБ Матерыялы па археалогii Беларусi МАИЭТ Материалы по археологии, истории и этнографии Таврии МИА Материалы и исследования по археологии СССР МЧК Могильники черняховской культуры ПАВ Петербургский археологический вестник ПИФК Проблемы истории, филологии, культуры РА Российская археология РСМ Раннеславянский мир СА Советская археология САИ Свод археологических источников СПб Санкт-Петербург СГЭ Сообщения Государственного Эрмитажа ТГИМ Труды Государственного исторического музея ТГЭ Труды Государственного Эрмитажа ХСб Херсонесский сборник

6 6 Jacek ANDRZEJOWSKI, Renata MADYDA-LEGUTKO Bronze Belt Buckles with Doubled Tongue between Scandinavia and the Black Sea. In search of local and inter-regional connections during the Roman period 1 Within the area of the Central-European Barbaricum, in phase B2b of the Early Roman period and at the beginning of the Younger Roman period, 2 one of the characteristic elements of the warrior s clothing were relatively wide belts fastened with massive rectangular double-tongue belt buckles. Such buckles constitute a rather varied group in terms of frame construction (uni- or bi-partite), its proportions and sizes, shape of the tongue (fork-shaped or H-shaped), presence or lack of chape and ornamentation. They were discussed in detail by Renata Madyda-Legutko (1987; 1990), who classified them under types in group G. Considering the features of rectangular double-tongue belt buckles, known by 1990, two main groups have been distinguished: buckles with fork-shaped tongues (types 35 41) and buckles with H-shaped tongues (types 42 44). Separate, type 45, covers belt buckles with a tongue with distinctly widened base. Since that time we have observed an immense growth of finds, in particular from the archival sources that has been revealed recently, referring to the materials from the area of former East Prussia, as well as from inquiries in museums and recent excavations. Among the newly disclosed belt buckles, there are specimens with relatively small frames as well as with a different way of fitting the doubletongue than recorded so far. Also, the shape of the tongue base varies. It results in the necessity to modify the current system of classifying these belt buckles, but first and foremost, it changes the view on their distribution. While Madyda-Legutko registered about 105 specimens of double-tongue buckles, at the moment their number reaches almost 250 specimens. What is particularly essential is the two-and-ahalf increase in the number of such buckles from the sites of West Balt cultures, in particular from the relatively dense settlement of the Bogaczewo culture in Mazury today we know as many as over 110 specimens from the entire West Balt area, including about 30 specimens from two burial grounds: at Babięta, powiat (district) Mrągowo (former Babienten, Kreis [district] Sensburg), and Spychówko, powiat Szczytno (former Klein Puppen, Kreis Ortelsburg), both in Poland. 3 It requires us to try a different approach to the issue of distributing the pattern of the buckles with doubled tongue in the Barbaricum, recognized until now as almost ethnographic indicator of male clothing 1 We must say our special thanks to Mirosław Pietrzak and Małgorzata Tuszyńska, both from the Archaeological Museum in Gdańsk, who allowed us to publish the finds from Pruszcz Gdański 7, grave 149B, and to Marek Rubnikowicz, PhD, director of the District Museum in Toruń, for permission to publish the buckle from Papowo Biskupie. We also wish to express our gratitude to Anna Juga-Szymańska, PhD, from the University of Warsaw, Institute of Archaeology, for her assistance in inquiry in the personal files of Martha Schmiedehelm, and to Agata Chilińska-Drapella, temporarily from the Zentrum für Baltische und Skandinavische Archäologie in Schleswig, who helped us to reconstruct a grave assemblage from Sovkhoznoye. 2 We apply here the terms Younger and Late Roman periods introduced by Kazimierz Godłowski (1992a: 23, note 1). 3 M. Schmiedehelm (2011: 143) listed the finds only. More precise data come from archival sources (Schmiedehelm, personal files; Voigtmann, personal files; Jankuhn, personal files; Jakobson, personal files; PM Inventarbücher 7 & 10; cf. Bitner-Wróblewska et al. 2008: , 174, pl. LX CXXVII, CXCVIII CCIV; 2011: , ).

7 Bronze Belt Buckles with Doubled Tongue between Scandinavia and the Black Sea. In Search of Local and Inter-Regional Connections during the Roman Period in the Przeworsk culture from late stage of phase B2 and phase B2/C1 C1a (e. g. Madyda-Legutko 1987: 85 86, 91; 1990; cf. Nowakowski 1991: 54; 1994: 374). As of now, at the whole vast area of the Przeworsk settlement from the turn of Early and Younger Roman periods there have been 80 such belt buckles registered. At other areas doubletongue belt buckles are much less frequent. The majority of the double-tongue belt buckles found within the Central-European Barbaricum are relatively large specimens, whose bi-partite frame was set on the bar. Such construction guaranteed certain flexibility of the fastening itself. It is highly possible that originally, all such belt buckles were fitted with a chape, which also ensured that it is strongly fastened to the strap. Belt buckles with uni-partite rectangular frame with a fork-shaped tongue, that is type AG35, are mostly small- and medium-sized, however, a few larger specimens are known as well (up to ca. 80 mm; Oblin, powiat Garwolin, Poland, grave 193 Czarnecka 2007: 51, pl. CLXXIX / ). Among them there are buckles with relatively small rectangular frame and similar, also small specimen with almost square as well as slightly trapezoid frame. Similar uni-partite buckles equipped with a chape represent type AG36. They are very different in size the height of the smallest specimen is only 25 mm (Asdrup, Denmark see: List 1), while the largest one extends to 108 mm in height (Stryczowice, Poland see: List 2). Most of the double tongue rectangular belt buckles belongs to type AG37 bi-partite buckles with a chape and a fork-shaped tongue; their height varies from about 31 mm (Badow, Germany, grave 658 see: List 3) to 110 mm (Janowo, powiat Mrągowo [former Heinrichsdorf, Kreis Sensburg], Poland, grave 170 Schmiedehelm, Personal files, 7.22b 108). This type includes also several buckles preserved without chape, previously defined as type 41, and some buckles with elaborated decoration. Among them there are iron buckles with silver inlay (Oblin, Poland; Badow, Germany; perhaps also Hagenow, Germany, grave 5/1995 see: List 3) and bronze buckle with open-work chape (Witaszewice, Poland see: List 1). The buckles with H-shaped tongue are significantly larger than buckles with fork-shaped tongue (medium height ratio ca. 83 mm to ca. 68 mm), however, they are notably less numerous (ca. 59 to ca. 148). Among uni- and bi- partite buckles with such tongue (types AG42 and AG43 respectively) there are also some remarkably smaller specimens, like the bronze buckle from Tårnholm, Denmark (height ca. 38 mm see: List 1), the iron buckle with silver inlays from Hagenow, Germany, grave VII/1905 (height ca. 44 mm see: List 3), and the iron specimen from Kovrovo, Kaliningrad oblast, Russia (former Dollkeim, Kreis Fischhausen), grave 35 (height 28 mm) individual traits of the latter (Raddatz 1957: 38; 1993: fig ; Madyda- Legutko 1990: 585; Nowakowski 1996: 18, pl ; Кулаков 2004: 18, fig ) place it, however, slightly apart the type. The largest known buckle with H-shaped tongue, i. e. an iron specimen with open-work chape from Chmielów Piaskowy, powiat Ostrowiec Świętokrzyski, Poland (stray find), reaches 116 mm in height (Godłowski, Wichman 1998: 44 45, pl. LXXXVII. 4; XCIII. 4). Assorted uni- and bi-partite buckles with elongated base of the tongue form a separate group of buckles with doubled tongue (type 45). Tongues of these buckles are mostly fork-shaped, with slightly widened base; in some specimens this feature is very poorly marked (e. g. Arbasy, powiat Siemiatyczne, Poland, grave 121 unpublished, personal communication from Katarzyna Rusin), and the borderline between such buckles of type 45 and some other specimens (like e. g. buckle type 37 from Kraśnik-Piaski, powiat Kraśnik, Poland, grave 11 Wichrowski 2004: 318, fig ), remains unclear. However, the type includes also specimens with the base of fork- or H-shaped tongue elongated towards the sides of the frame; the base itself is either narrow or almost as broad tas the frame (cf. Szymański 2005: 31 33, fig ). Beyond the presented classification stays a few unique rectangular or sub-rectangular belt buckles. Three specimens with both uni- (Opatów, Starachowice, both in Poland) and bipartite frames (Kamień, Poland) are equipped with tongues fitted on an internal bar. 4 Two other buckles with two loosely fitted separate tongues (Lazovskoye, Russia; Żdżarów, Poland), 5 basing on their stylistic and morphological characteris- 7 4 Uni-partite: Opatów, powiat Kłobuck, Poland, stray find (Madyda-Legutko, Rodzińska-Nowak, Zagórska-Telega 2011a: 382 [here as modern buckle]; 2011b: pl. CDXVII. 237); Starachowice, Poland see List 3. Bi-partite: Kamień, powiat Pisz, Poland (former Kreis Sensburg), grave 2 (Juga, Ots, Szymański 2003: 214; Szter 2010: 253, pl. I. 3). 5 Lazovskoye see List 1; Żdżarów, powiat Sochaczew, stray find Nowakowski 2003: 306, pl. LV. 22.

8 8 Jacek Andrzejowski, Renata Madyda-Legutko tics, may be dated to the beginning of the Late Roman period. They are similar, to some extent, to the silver buckle of an ornamental belt from Neudorf-Bornstein, Germany, well-dated to phase C2, as well as to the iron buckle from Westerwanna, Germany. 6 Within the whole area where they are recorded, the rectangular double-tongue belt buckles of types are mostly iron specimens, 7 bronze buckles are relatively rare out of the total number of almost 250 specimens, only a little over 30, that is little over 13 %, were produced using this material (see: List 1). As a consequence, it is the non-iron belt buckles that raise particular interest. The scope of the rectangular double-tongue belt buckles made of bronze is clearly distinctive when compared with the scope of the iron specimens. In particular, what is important here is almost the exclusive presence of bronze belt buckles within the area of the Wielbark culture, where there is only one iron specimen recorded a fork-shaped tongue of a belt buckle was to be found in Elbląg-Pole Nowomiejskie, Poland (former Elbing-Neustädterfeld; Schmiedehelm 2011: 145). Recently an iron buckle type AG36 and an iron fork-shaped tongue have been registered at an Early Roman period cemetery at Kołat, powiat Lipno, Poland (Sobczyk 2004: 22, fig ; 2011: 347, fig. 8. 8; and personal communication from Paweł Sobczyk). However, this site is located very close to the territories settled by populations of the Przeworsk culture. Bronze belt buckles are rather unique within West Balt areas 8 (only five out of over 110 specimens) and within the Przeworsk culture area (only four out of 80 specimens), not even one was found to the south of the Carpathians, despite the presence of a few iron specimen within that area (fig. 1). 9 Bronze belt buckles are also rare in the Elbian zone, while they constitute a significant percentage in Jutland and on Danish islands (five out of 14 specimens). The rectangular doubletongue belt buckles found in the northern Black Sea area are almost exclusively bronze; here (Kerch, Ukraine) a totally unique silver belt buckle, alike type AG45 was found (Martin 1894: fig. 62); the tongue of a bronze belt buckle from Divochky, Ukraine should probably be associated with this group. According to archaeological indicators, bronze double-tongue belt buckles, just as iron specimens, were characteristic of male attire (Asdrup, Denmark; Prusiek, Podwiesk, grave from 1876, and Spychówko all in Poland; Sovkhoznoye, Russia 10 ; Olbia, Ukraine; Witaszewice, Poland; probably also Thorsberg, Germany, and maybe Kerch, Ukraine, too see List 1). Some other burials (Kholmskoye, Ukraine; Kowalewko, Prusiek, and Pruszcz Gdański all in Poland) are confirmed as male by anthropological analyses only (Сегеда, Дячeнко 1984: 96; Rewekant, Segieda 2001: 428; Gładykowska-Rzeczycka et al. 2003: 345; personal communication from Joanna Zagórska-Telega). This rule is not so strict, as in some cases the grave equipments indicate that even large bronze belt buckles, corresponding to the iron belt buckles of the so-called sword belts, could be elements of female attire. We are talking here about three or four assemblages from the area of the Wielbark culture: from Ruże, Jelonki, Weklice, grave 448 all 6 Neudorf-Bornstein, Kreis Rendsburg-Eckernförde in Schleswig, Germany, grave 7 (Carnap-Bornheim 2002: 23, pl. 1; 2003: 242, fig. 2); Westerwanna, Landkreis Cuxhaven in Lower Saxony, grave 1355 (Zimmer-Linnfeld 1960: pl b). 7 A few iron buckles were decorated with silver inlays (see: List 3; cf. Madyda-Legutko 1990: ) or bronze rivets (see: List 2; fig ). The latter were probably more numerous, however, bronze elements might totally disappear in fire of a cremation pyre. Perhaps it was the case of well-preserved iron buckle from a cremation grave from Drozdowo, Poland (see: List 2), with ten rivet holes, but none of the rivets remained. 8 In her PhD thesis from 1944, M. Schmiedehelm (2011: ) listed all buckles with fork- and H-shaped tongue from known at the time, however, she stated that only one specimen from the areas settled by Western Balts was made of bronze, what does not correspond with her own personal files. 9 Madyda-Legutko 1990: fig. 1. Next finds come from Bohemian Basin (Třebusice, powiat Kladno; Modřice, okres Brno, both in Czech Republic unpublished, personal communication from Eduard Droberjar and Jaroslav Tejral respectively) and from the Hungarian plain (Gyulavári, comitat [county] Bekes Vaday 1987: 73, fig. 1. 5; Hévízgyörk, comitat Pest Dinnyés 1991: 154, pl ; Nyíregyháza-Felsősima, comitat Szabolcs-Szatmár-Bereg unpublished, personal communication from Eszter Istvánovits). 10 An eye brooch of Prussian series published by Oleg Radyush and Konstantin Skvortsov (Радюш, Скворцов 2008: fig. 8) as a part of grave 10 equipment in fact does not belong to this assemblage (personal communication from Agata Chilińska-Drapella).

9 Fig. 1. Distribution of bronze belt buckles with rectangular frame and fork- or H-shaped tongue in the Barbaricum (green stars) against distribution of similar iron buckles (red points). Numbers fit with the list of finds (List 1)

10 10 Jacek Andrzejowski, Renata Madyda-Legutko in Poland, 11 and perhaps also Leśno, Poland 12 (see: List 1). Next burial with female equipment comes from the West Balt territory, however, the doubletongue bronze buckle from grave 6 at Sławosze, Poland is hardly known. A relatively small bronze belt buckle from the grave with definitely female equipment from Kemnitz, Germany in the light of the anthropological analysis, belonged to the burial of probably a child aged infans I (Geisler 1974: 42; 1984, p. 163). The chronology of the bronze double-tongue belt buckles mostly corresponds to the dating of the iron specimens, that is between the late stage of phase B2 of the Early Roman period and stage C1a of the Younger Roman period. However, the majority of the belt buckles from the assemblages with reliable dating comes from stage B2/C1 C1a. The exceptions are: the buckle from Kemnitz mentioned above, the buckle from grave 139 from Weklice, Poland and in particular the buckle from grave 149B from Pruszcz Gdański 7, Poland. While the large belt buckle from Weklice, Poland comes from a grave dated at the downturn of B2 phase (cf. Natuniewicz-Sekuła, Okulicz-Kozaryn 2011: 52, 127, pl. LIV. 139), the other two specimens may be apparently older. The grave from Kemnitz contained a silver brooch Almgren 29 and a pair of bronze brooches Almgren 153 (Geisler 1974: 42, pl ), which let us place the assemblage in phase B2b. Particular attention is drawn to a beautiful set of bronze fittings of a belt from grave 149B from Pruszcz Gdański, with a belt buckle type AG37 (fig. 2), which may be the oldest reliably dated double-tongue belt buckle in the Central- European Barbaricum. A bronze brooch type Almgren 60 (fig. 2. f) found in the grave is recognized as an indicator of an early stage of phase B2, yet in the light of the more recent studies on eye brooches of Prussian series, their chronology within the Wielbark culture area should be extended also to the younger stage of the phase, that is stage B2b (recently Mączyńska 2004: , tab. 1; Cieśliński 2010: 55). The decorative pattern on the crease of the chape (fig. 2. a) refers to the patterns known from e. g. Norican-Pannonian belt clasps (cf. Garbsch 1965: fig , and plates), which would suggest its earlier (i. e. stage B2a) origins; such dating fits with dating of five profiled bronze end-fittings (fig. 2. b) alike the Przeworsk types 1/3 and 1/4 (cf. Madyda-Legutko 2011: 17 20, 23 24, pl. I. 4 11, 13 15; IV. 1 13) and is confirmed by dating of a grave 149A, dug into the pit of grave 149B (personal communication from M. Pietrzak). Finally, this belt buckle should be recognized as the oldest known specimen of double-tongue belt buckles in the Central-European Barbaricum, as none of the iron belt buckles AG35 45 types cannot be dated reliably earlier than younger stage of phase B2 (Madyda-Legutko 1990: ). 13 The belt buckle from Pruszcz Gdański, grave 149B, corresponds chronologically to the few specimens of semi-circular belt buckles with uni-partite frame and fork-shaped tongue, type AD4, known from Central Europe; among them there is also a single bronze specimen. 14 Bronze double-tongue belt buckles are a pretty varied group, both in terms of construction and style. Explicit individualisation of particular specimens is probably the result of the raw material characteristics, enabling a much freer modelling of the final product than it was possible in case of iron specimens. As a result there are several belt buckles decorated with a composed profiled ornament (e. g. fig. 3. 2, 6, 14; 4. 2). The use of bronze, giv- 11 Female indicator (three brooches) is strong enough to rebut results of anthropological analysis (an adult male Natuniewicz-Sekuła, Okulicz-Kozaryn 2011: 112). The grave was partly damaged by another burial (grave 452), identified as female both on archaeological and anthropological means (see: Natuniewicz-Sekuła, Okulicz- Kozaryn 2011: , pl. CXCVIII CCI). 12 K. Walenta (2009: 16) recognizes burial 1 in grave II as male, however, basing only on a male type of the buckle. The grave was robbed but two glass beads preserved in the pit (Walenta 2009: 14, pl. XIII. 1, 2) suggest a female burial. 13 Aforementioned iron buckle type AG36 from grave 2 at Kołat, Poland was found together with bronze endfitting alike Przeworsk type 1/6 (Sobczyk 2011: fig. 7. 7; 8. 8; cf. Madyda-Legutko 2011: 18 19, 24 25, pl. II, V. 1 9), dated broadly to phase B2. 14 Bronze buckle was found at Tvršice, okres Louny, Czech Republic (Motyková-Šneidrová 1965: 131, fig ; Madyda-Legutko 1987: 26, pl. 8). Except for a buckle from Tarnów, powiat Opole, Poland, grave 105/1938, quoted by Madyda-Legutko (1987: 26) there are also iron buckles from Brudnice, powiat Żuromin, Poland, grave 72 (Szela 2009: fig ), Stupsk, powiat Mława, Poland, grave 3/89 (unpublished; collection of the Muzeum Ziemi Zawkrzeńskiej in Mława), and Degucie, powiat Gołdap, Poland (former Zapfengrund, Kreis Goldap; Engel 1939: 53, fig. 8. d; Schmiedehelm, personal files: ).

11 Bronze Belt Buckles with Doubled Tongue between Scandinavia and the Black Sea. In Search of Local and Inter-Regional Connections during the Roman Period 11 Fig. 2. Bronze eye-fibula (f) and set of bronze belt fittings from inhumation grave 149B from Pruszcz Gdański 7: buckle type AG37 (a), end-fittings type Raddatz O.18 (b), box-like mountings covered with silver sheet (c), belt hangers (d, e). (Drawing: A. Kruszlińska; courtesy of the Muzeum Archeologiczne, Gdańsk)

12 12 Jacek Andrzejowski, Renata Madyda-Legutko ing them a spectacular, golden colour, emphasized the ceremonial nature of the belt with fittings made of this material. Among the decorated belt buckles, the specimens from Kowalewko (fig ) and Witaszewice (fig. 4. 1), both in Poland, stand out. The base of the tongue and frame and the edge of the chape of the belt buckle from Kowalewko are decorated with imitation of notched wire inlay. On the frame of the belt buckle from Witaszewice there is a row of concentric circles and a zig-zag line, and the edge of the chape is decorated with an openwork stripe with the pattern comprising of arcs and rhombi. Openwork ornamentation in the Younger Roman period is present at metal elements of the belts work in male attire, mostly in the West Balt area, less frequently in the Przeworsk culture (recently see: Madyda-Legutko 2009). The bucle from Witaszewice was found together with an openwork plate with four-leaf motif, possibly an element of a belt set (Kaszewska 1971: ; cf. Madyda-Legutko 1990: 572, fig. 9. 1; 2009: 778, pl. 3. 1b). On the one hand, this motif is a parallel to some finds from the Younger Roman period from the foreground of the Danubian limes, on the other hand to the finds from the North-European bog deposits (Madyda-Legutko 2011: 118, note 360). The bronze double-tongue belt buckles are dominated by specimens with distinctly rectangular frames, two to three times larger in height than their width, bi-partite, that is with a frame set on the bar, usually equipped with a narrow, rectangular chape. Among the belt buckles with a bi-partite frame, the most frequent ones are the specimens with fork-shaped tongue, resembling type 37 (fig ; ). They mainly come from the scope of the Wielbark culture settlement (Kowalewko; Leśno; Papowo Biskupie; Podwiesk, grave from 1876; Pruszcz Gdański; Ruże; Weklice, grave 448 all in Poland), 15 and individual specimens are known from the Przeworsk culture (Adolfin, Poland), from Funen (Sadderup, Denmark, chapeless), Schleswig (possibly Thorsberg) and northern Black Sea zone (Kholmskoye). Some specimens of these belt buckles, due to their individual characteristics, were previously distinguished as types AG39 (Witaszewice), AG40 (Kemnitz) and AG41 (Czarne, Poland). All the well preserved specimens of bronze fork-tongue belt buckles have rectangular chapes bound with a few rivets, which seems to be justified by practical reasons while fixing them to relatively wide belts; in case of the belt buckle from Czarne, the function of the chape could be realized by a rectangular plate with two rivets (cf. Eggers, Stary 2001: ). Less frequent than the discussed ones are bronze bi-partite belt buckles with H-shaped tongue, equipped with two grips to place along the belt buckle bar, corresponding to type AG43 (fig ). They are present in the Wielbark culture area (Weklice, grave 139; chapeless buckle from Jelonki; supposedly also Wyszków all in Poland), in the Samland peninsula (Sovkhoznoye, Russia; probably also unprovenanced chapeless buckle from East Prussia) and in south-eastern reaches of the Przeworsk culture territory (Prusiek, Poland). The belt buckle with a slightly rounded 15 Perhaps also unprovenanced buckle from the collection of the State Archaeological Museum in Warsaw. Fig. 3. Bronze belt buckles with uni- (1 5) or bi-partite frame (6 16) and fork- (1 4, 6 16) or H-shaped (5) tongue. 1 Kerch / Керчь (Vasil yev 2005); 2 Olbia, vault 32/1910 (Бажан, Герцегер 1993); 3 Divochky / Дівочки, stray find (courtesy of Maxim Levada); 4 Asdrup, grave (drawn from the NMK inventory book); 5 Tårnholm, stray find (drawing: Jacek Andrzejowski); 6 Sadderup, stray find (drawing: Andrzejowski); 7 Ruże, grave 8 (Romanowska-Grabowska 1972); 8 Leśno I, grave II.1 (Walenta 2011); 9 Thorsberg, bog find (Raddatz 1957); 10 Papowo Biskupie, stray find (drawing: B. Bielińska-Majewska, courtesy of Muzeum Okręgowe, Toruń); 11 Unprovenanced, collection of Państwowe Muzeum Archeologiczne, Warszawa (drawing: Andrzejowski); 12 Czarne, grave (Eggers, Stary 2001); 13 Kemnitz, grave 356 (Geisler 1974); 14 Kowalewko, grave 369 (Skorupka 2001); 15 Podwiesk, grave (Kleemann 1938); 16 Podwiesk, grave from 1873 (Jankuhn, personal files)

13 Bronze Belt Buckles with Doubled Tongue between Scandinavia and the Black Sea. In Search of Local and Inter-Regional Connections during the Roman Period 13

14 14 Jacek Andrzejowski, Renata Madyda-Legutko

15 Bronze Belt Buckles with Doubled Tongue between Scandinavia and the Black Sea. In Search of Local and Inter-Regional Connections during the Roman Period frame from Lazovskoye in Samland, Russia, with two loose tongues stabilized by a chape, is probably their alternation (fig. 5. 5). Specimens with uni-partite frames are sparse. Two such buckles discovered in Kerch (fig. 3. 1) and Olbia (fig. 3. 2), both in Ukraine, with a relatively narrow, high frame along with a chape and fork-shaped tongue, correspond with type AG36; perhaps the fork-shaped tongue from Divochky (fig. 3. 3) belongs to the same type. Two small buckles from western Scandinavia, from Asdrup (fig. 3. 4) and Tårnholm (fig. 3. 5), both in Denmark, refer to type AG35/36 and AG42 respectively, however, both do not find exact analogies among iron buckles from the Central-European Barbaricum. Among bronze belt buckles there are also bipartite rectangular specimens with a chape, having a broader base of the double-tongue, similar to type AG45. They include the fork-shaped-tongue belt buckle from a bog deposite in Thorsberg, Germany with broader base until the internal edge of the frame (fig. 5. 2) and the damaged belt buckle from Rauschendorf, Germany with H-shaped tongue, also with a broader base (fig. 5. 4); the latter seems to be comparable with an iron specimen found at Brokær mark, Ribe AMT, in Jutland, Denmark (Rasmussen 1995: 52, fig. 12. c). The tongue of the bronze belt buckle from Spychówko, Poland (fig. 5. 1) and separate bronze tongue from Jakuszowice, Poland (fig. 5. 3) resemble type AG45, namely its variety with only slightly broader tongue base, similar to tongues of iron belt buckles from e. g. former Koddien, Kreis Wehlau, in Samland, Russia (Szymański 2005: 33, note 63, fig. 5. 6), and above mentioned grave 121 from Arbasy in eastern Poland. Tongues with broader base are typical 15 for belt buckles, also for single-tongue belt buckles, known mainly from the West Balt area, both in the Bogaczewo and Dollkeim-Kovrovo cultu res (e. g. Szymański 2005: 31 33, fig. 5). However, a decorative, rectangular bronze belt buckle with widened base of a single tongue, decorated with silver inlays, comes also from the Wielbark culture area (Leśno 1/22, powiat Chojnice, Poland, stray find Madyda-Legutko 1987: 51, 152, pl ; Walenta 2009: 10, pl. V. B; XCIX. 2). Particular attention should be drawn to a silver belt buckle from Kerch, with bi-partite, almost square frame and double-tongue with a distinctively broader, rectangular base, filling the inner space of the frame, having a rectangular chape bound with five rivets (fig. 5. 6). In this case the attention is drawn by the material used, which was almost never used to make rectangular belt buckles with double-tongue. 16 However, aforementioned silver specimen from Kerch resembles with proportions of the frame and distinctly broader tongue base some Crimean bronze belt buckles with quadrangular, uni-partite frame and single-tongue with widened base which chapes are bound by five rivets located in the corners and in the centre (e. g. Васильев 2005: fig. 1. 3; 2. 1, 10; Шаров 2010: fig. 6. H; Vasil yev 2011: fig , 13). The listed types of bronze rectangular doubletongue belt buckles have their counterparts among iron specimens. Only the unique uni-partite or bipartite belt buckles which double-tongue is set on an internal bar (see note 4 above; cf. similar buckles with single tongue, types AG27 and AG50 Madyda-Legutko 1987: 51, 56, pl ; ), do not have their bronze matches yet. 16 Unique silver bi-partite belt buckle with two loose tongues belongs to the set of mountings of a ceremonial belt from grave 7 from Neudorf-Bornstein, Kreis Rendsburg-Eckernförde in Schleswig, Germany, dated to phase C2 (Carnap-Bornheim 2002: 23, pl. 1; 2003: 242, fig. 2). Fig. 4. Bronze belt buckles with bi-partite frame and fork- (1 5) or H-shaped (6 11) tongue. 1 Witaszewice, grave 22 (Kaszewska 1971); 2 Pruszcz Gdański 7, grave 149B (drawing: A. Kruszlińska); 3 Adolfin, grave 1 (Zielonka 1961); 4 Sovhoznoye, grave 10 (Jankuhn, personal files); 5 Kholmskoye / Холмскoе, grave 16 (Гудкова, Фокеев 1984); 6 Wyszków, stray find (drawing: A. Potoczny); 7 Jelonki, Fund 1 (Nowakowski 1994); 8 Unprovenanced, Eastern Prussia (Schmiedehelm, personal files); 9 Weklice, grave 139 (Natuniewicz-Sekuła, Okulicz-Kozaryn 2011); 10 Prusiek, grave 40 (scanning and drawing: U. Socha); 11 Weklice, grave 448 (Natuniewicz-Sekuła, Okulicz-Kozaryn 2011)

16 16 Jacek Andrzejowski, Renata Madyda-Legutko

17 Bronze Belt Buckles with Doubled Tongue between Scandinavia and the Black Sea. In Search of Local and Inter-Regional Connections during the Roman Period Bronze, rectangular belt buckles with doubled tongue, like the iron specimens, in the Barbaricum were combined with end-fittings creating sets that were characteristic for male clothing. The design of these sets is not uniform, which should be considered a result of stylistic changes that took place while using double-tongue belt buckles for clothing. Bronze belt set from an oldest known assemblage with rectangular buckle with doubletongue (older stage of B2 phase) discovered at the Wielbark culture cemetery in Pruszcz Gdań ski 7, grave 149B, comprises of, apart from the belt buckle, five end-fittings with profiled shaft and a few mountings sliding on the length of the belt arched belt hanger, two silver-plated box mountings and three links for a narrow leather strap probably suspend at the belt (fig. 2. a e). The se cond bronze set, comprising of two end-fittings with profiled shaft, comes from Kemnitz, grave 356, in Brandenburg, from a grave dated to B2b stage. A belt set from Pruszcz Gdański, with significant number of elements, is close in style to the belt set found in the Przeworsk cemetery at Kamieńczyk, powiat Wyszków, Poland, grave 98, that comprises of an iron rectangular belt buckle with fork-shaped tongue and a few bronze profiled end-fittings, as well as iron and bronze hangers for the length of the belt (Dąbrowska 1997: 29, pl. LII ). Both sets refer to the decoration of the belts fastened with belt buckles with various frames, with a higher number of end-fittings with profiled shafts and with hangers for the length of the belt, used to hang the so-called personal equipment. Such belts are known from the Przeworsk culture (e. g. Marzęcin, powiat Pińczów, Poland), Wielbark culture (e. g. Leśno 2, powiat Chojnice, Poland, graves 12 and 81), as well as from western Mecklenburg (e. g. Hagenow, Kreis Ludwigslust, grave 9/1995; Marwedel, Kreis Lüchow-Dannenberg, grave I both in Germany); these assemblages are dated 17 to the beginning of B2 phase (for references see: Madyda-Legutko 2011: ). In the younger stage of B2 phase and in C1a phase, belts fastened with bronze belt buckles with doubled tongue, just as iron specimens, were fitted with end-fittings ending with a ring (Jelonki; Podwiesk, grave from 1876 both in Poland; Sovkhoznoye) or a ring with a protrusion (Prusiek; Weklice, grave 139 both in Poland), and flat sub-rectangular fittings (Kowalewko; Wyszków both in Poland). Significant distribution of wide belts fastened with such buckles and massive endings, typical for a warrior s attire, within the Central-European Barbaricum in phases B2b C1a (fig. 1), probably proves that the various groups of warriors, from various cultural environments, kept in close contacts with each other. In particular, it concerns the warriors from the Przeworsk culture and West Balt cultures of Bogaczewo and Dollkeim-Kovrovo, and possibly also from the Wielbark Culture. From the other hand, distribution of such belt-buckles made of bronze suggests that the Wielbark culture was the centre of their production and use. Perhaps most of the bronze double-tongue belt buckles from the territories of the iron Przeworsk culture as well as from West Balt zone was made in the Wielbark culture workshops. However, it is intriguing that until now double-tongue belt buckles are unknown from the Wielbark culture assemblages from phases B2/C1 C1a from the Northern and Eastern Mazowsze and Podlasie; the only exception may be the bronze belt buckle from Wyszków. From these areas, the double-tongue belt buckle pattern could reach the Elbian zone, Jutland and Danish Islands on the West, where it was probably reproduced in local workshops. Most likely this is the case of bronze double-tongue belt buckles from Danish Sadderup, Tårnholm, and Asdrup, while bronze buckles found in Thorsberg possibly came Fig. 5. Bronze (1 5) and silver (6) belt buckles with fork-shaped tongue with elongated base (1 5) or with two loose tongues (6), and iron belt buckles with doubled tongue and bronze rivets (7 9); bronze marked with yellow colour). 1. a, b Spychówko, grave 12 (a Schmiedehelm, personal files, b Szymański 2005); 2 Thorsberg, bog find (Raddatz 1957); 3 Jakuszowice, stray find (drawing: U. Socha); 4 Rauschendorf, grave (Geisler 1975); 5 Lazovskoye, stray find (Jankuhn, personal files); 6 Kerch / Керчь (Martin 1894); 7 Stryczowice, grave (Madyda-Legutko 1990); 8 Czacz, stray find (Kostrzewski 1956); 9 Chmielów Piaskowy, stray find (Madyda-Legutko 1990)

18 18 Jacek Andrzejowski, Renata Madyda-Legutko from the Wielbark or West Balt zone. The same may apply to bronze specimens from German Kemnitz 17 and Rauschendorf. Belt buckles with double tongue from the areas to the south of Carpathians are probably imported from the Przeworsk culture areas; it is more difficult to interpret individual, separate finds of only bronze belt buckles in the northern Black Sea area (cf. Васильев 2005; Шаров 2010: ). It seems most probable that these are imported items, connected with the penetration of the Pontic zone by the people of the Wielbark culture, which took over the vast areas of the former Eastern zone of the Przeworsk culture, reaching as far as Volhynia already in phase B2/C1 C1a (see e. g. Bierbrauer 1992: 17 20; Godłowski 1992b: 64 68; Русанова 1993; Щукин 2002; 2005: ). These finds are another proof of the information from written sources, saying that the Gothic tribes moved towards Oium lands (Jordanes Getica 25 28). 18 A semi-finished fork-shaped tongue from Divochky connects the belt buckles from the northern Black Sea area with the settlement zone of the Wielbark culture. 19 The same north-western direction of connections may be documented by the unique silver belt buckle found in Kerch. Appendix List 1. Bronze belt buckles of type AG Adolfin, powiat Aleksandrów Kujawski, Poland, grave 1. Type AG37. H. ca. 40 mm (fig. 3. 7) (Zielonka 1961: 198, fig. 2. b; Madyda-Legutko 1987: 53, 134). 2. Asdrup / Astrup, Skarresø sogn, Øster Lisbjerg herred, Randers amt, Denmark, grave. Type AG36. H mm (fig. 3. 4). Unpublished (Nationalmuseet København: C 25785; cf. Voß 2008: 271, list 1. 1; Blankenfeldt 2009: 454). 3. Czarne, powiat Gryfice (former Neuenhagen, Kreis Regenwalde), Poland, grave. Type AG37/41. H. 44 mm (fig ) (Kunkel 1935: 329, note 53; Madyda-Legutko 1987: 54, 139; 1990: 573, 584; Eggers, Stary 2001: 117, pl ). 4. Divochky / Дівочки, Chernyakhov district, Zhitomir oblast, Ukraine, stray find. Type AG36 (?) semiproduct of a fork-shaped tongue (fig. 3. 3). Unpublished (personal communication from Maxim Levada, Kiev). 5. Jakuszowice, powiat Kazimierza Wielka, Poland, stray find. Type AG45 fork-shaped tongue only (fig. 5. 3). Unpublished (Instytut Archeologii, Uniwersytet Jagielloński, Kraków). 6. Jelonki, powiat Elbląg (former Hirschfeld, Landkreis Elbing), Poland, Fund 1. Type AG43. H. 68 mm (fig. 4. 6) (Nowakowski 1994: 168, pl. VI. 11). 7. Kemnitz (Werder [Havel]-Kemnitz), Landkreis Potsdam-Mittelmark, Brandenburg, Germany, grave 356. Type AG37/40. H. 35 mm (fig ) (Kramer 1959: 283, pl. 44c. d; Geisler 1974: 42, pl. 32 / ; 1984: ; Madyda-Legutko 1987: 40, 111, pl. 16. AG40; 1990: fig. 3. 1; 7. 4). 8. Kerch / Керчь, Autonomous Republic of the Crimea, Ukraine, stray find. Type AG36. H. ca. 78 mm (fig. 3. 1) (Васильев 2005: 279, fig. 1. 1; Vasil yev 2011: 136, fig ). 9. Kholmskoye / Холмскoе, Khartsyzsk district, Odessa oblast, Budzhak, Ukraine, grave 16. Type AG37. H. ca. 65 mm (fig. 4. 5) (Гудкова, Фокеев 1984: 16, fig. 6. 2; Madyda 1990: 584). 10. Kowalewko, powiat Oborniki, Poland, site 12, grave 369. Type AG37. H mm (fig ) (Skorupka 2001: 99, pl. 115 / ). 11. Lazovskoye / Лазовское, Kaliningrad oblast (former Trömpau, Kreis Königsberg), Russia, stray find. Unique bi-partite buckle with two loose tongues. H. 76 mm (fig. 5. 5). Unpublished (Jankuhn, personal files). 12. Leśno, powiat Chojnice, Poland, site 1/22, grave II/1. Type AG37. H. 62 mm (fig. 3. 8) (Walenta 2009: 14, pl. XIII. 3 here as type AG36). 17 A. Leube (1975: 26), and after him H. Geisler (1984: 112) believe this buckle is a local product basing on eastern (Przeworsk) pattern. 18 For the history of the Gothic tribes see: (Wolfram 2009). 19 It seems unlikely that the bronze belt buckles with doubled tongue from of the Pontic zone were the result of Sarmatian influences from the Great Hungarian plain the only three known there belt buckles of this type are the non-decorated iron specimens (cf. note 9).

19 Bronze Belt Buckles with Doubled Tongue between Scandinavia and the Black Sea. In Search of Local and Inter-Regional Connections during the Roman Period Olbia, Nikolayev oblast, Ukraine, burial vault 32/1910. Type AG37. H. ca. 60 mm (fig. 3. 2) (Бажан, Герцегер 1993: 107, fig. 1. 1; Васильев 2005: 279, fig. 2. 9; Vasil yev 2011: 136, fig 2. 17). 14. Papowo Biskupie, powiat Chełmno, Poland, stray find. Type AG37. H. 64 mm (fig ). Unpublished (Janikowski 1976: 72; Madyda-Legutko 1987: 53, 160; Muzeum Okręgowe in Toruń: MT/A/775). 15a. Podwiesk, powiat Chełmno (former Podwitz, Kreis Kulm), Poland, grave. Type AG37. H. ca. 98 mm (fig ). Unpublished (Schmiedehelm 2011: 145; Schmiedehelm, personal files: ; cf. Lissauer 1887: 147). 15b. Podwiesk, powiat Chełmno (former Podwitz, Kreis Kulm), Poland, grave from Type AG37. H. ca. 75 mm (fig ) (Blume 1912: 53, note 2; Kleemann 1938: 28, pl. V. v; Madyda-Legutko 1987: 53, 161; Schmiedehelm 2011: 145; Schmiedehelm, personal files: ; Jahn, personal files). 16. Prusiek, powiat Sanok, Poland, grave 40. Type AG43. H. ca. 80 mm (fig ) (Madyda-Legutko, Rodzińska- Nowak, Zagórska-Telega 2007: 65; Uniwersytet Jagielloński in Kraków, Instytut Archeologii). 17. Pruszcz Gdański, powiat Pruszcz Gdański, Poland, site 7, grave 149B. Type AG37. H. 41 mm (fig. 2. a; 4. 2). Unpublished (Archaeological Museum in Gdańsk, personal communication from Mirosław Pietrzak and Małgorzata Tuszyńska). 18. Rauschendorf (Sonnenberg-R.), Landkreis Oberhavel, Brandenburg, Germany, grave. Type AG45. H. ca. 40 mm (?) (fig. 5. 4) (Geisler 1975: 192, fig. 1. b; Madyda-Legutko 1990: 554, fig here as undefined type). 19. Ruże, powiat Golub-Dobrzyń, Poland, grave 8. Type AG37. H ca. 70 mm (fig. 3. 7) (Romanowska-Grabowska 1972: fig. 6; Romanowska-Grabowska, Janikowski 1982: 108, pl. X. 3 here as a hinge mounting of a wooden chest or casket; Madyda-Legutko 1987: 164 not mentioned as made of bronze, cf. 53). 20. Sadderup / Såderup, Kullerup sogn, Vindinge herred, Svendborg amt, Funen, Denmark, stray find. Type AG37/41 (or AG37/40, chapeless). H. ca. 45 mm (fig. 3. 6). Unpublished (Nationalmuseet København: C 13499). 21. Sławosze (Święty Kamień-S.), powiat Kętrzyn (former Henriettenfeld, Kreis Gerdauen), Poland, grave 6. Type unknown (Heydeck 1879: 19; Schmiedehelm, personal files: ; 7. 13d 393; Madyda-Legutko 1990: 585). 22. Sovkhoznoye / Совхозное (Mayskiy), Kaliningrad (former Groß Friedrichsberg, Kreis Königsberg), Russia, grave 10. Type AG43. H 67 mm (Fig. 4: 9) (Schmiedehelm 2011: 145 here as made of iron; Raddatz 1999: 189, fig ; Радюш, Скворцов 2008: fig ; Grenz, personal files; Jankuhn, personal files). 23. Spychówko, powiat Szczytno (former Klein Puppen, Kreis Ortelsburg), Poland, grave 12. Type AG45. H. ca. 80 mm (fig. 5. 1a, 1b) (Schmiedehelm 2011: 143 here as made of iron; Schmiedehelm, personal files: a; cf. Szymański 2005: 33, note 63, fig , 11; personal communication from Christina Reich, Museum für Vor- und Frühgeschichte Staatliche Museen zu Berlin). 24. Tårnholm, Tårnborg sogn, Magelse herred, Sorø amt, Zealand, Denmark, stray find from a Bronze Age barrow. Type AG42 (chapeless). H. ca. 38 mm (fig. 3. 5). Unpublished (Nationalmuseet København: 6926). 25a. Thorsberg, Kreis Süderbrarup, Angeln (Schleswig-Holstein), Germany, bog find. Type AG45. H. 70 mm (fig. 5. 2) (Engelhardt 1863: 83, pl ; Raddatz 1957: 37, pl. 1. 9; Madyda 1990: 585, fig here as type AG45A; Blankenfeldt 2009: 29. P61; pl. 10. P61). 25b. Thorsberg, Kreis Süderbrarup, Angeln (Schleswig-Holstein), Germany, bog find. Type AG37/41. H 81 mm (Fig. 3: 9). (Raddatz 1957: 41, pl. 1: 10; Madyda 1990: 585 [here as undetermined type]; Blankenfeldt 2009: p. 29: P62, pl. 10: P62). 26a. Weklice, powiat Elbląg, Poland, grave 139. Type AG43. H mm (fig ) (Natuniewicz-Sekuła, Okulicz-Kozaryn 2011: 52, pl. LIV / ; CCXXXIX. 2). 26b. Weklice, powiat Elbląg, grave 448. Type AG37/40. H. 63 mm (fig. 4. 4) (Natuniewicz-Sekuła, Okulicz- Kozaryn 2011: 112, pl. CXCVII / ; CCXXXIX. 4). 27. Witaszewice, powiat Łęczyca, Poland, grave 22. Type AG37/39. H. 73 mm (fig. 4. 1) (Kaszewska 1971: pl. 167/2. 11; Madyda-Legutko 1987: 53, 172, pl. 16. G39; 1990: 5, 72, fig. 2. 6; 9. 1; 2009: 778, pl. 3. 1a). 28. Wyszków (Plażowa Street), powiat Wyszków, Poland, stray find. Type AG43. H. 69 mm (fig. 4. 6). Unpublished (Państwowe Muzeum Archeologiczne, Warszawa: IV / 10487: 1). 29. Unprovenanced, former East Prussia (perhaps Samland). Type AG43 (chapeless). H. ca. 72 mm (fig. 4. 8). Unpublished (Schmiedehelm 2011: 145; Schmiedehelm, personal files: ; personal communication from Anna Juga-Szymańska). 30. Unprovenanced. Type AG37/41. H mm (fig ). Unpublished (National Archeological Muzeum in Warsaw: IV/5637).

20 20 Jacek Andrzejowski, Renata Madyda-Legutko List 2. Iron belt buckles type AG35 45 decorated with bronze rivets 1. Chmielów Piaskowy (Chmielów), powiat Ostrowiec Świętokrzyski, Poland, stray find. Type AG43/39. H. 116 mm (fig. 5. 9) (Godłowski, Wichman 1998: 44, pl. LXXXVII. 4; XCIII. 4; cf. Madyda-Legutko 1987: 53 here as made of bronze, 137; 1990: fig. 2. 5; 2009: 778, pl. 3. 2a). 2. Czacz, powiat Kościan, Poland, stray find. Type AG43. H. 100 mm (fig. 5. 8) (Kostrzewski 1956: 76, no. 55, fig. 9. 6; Madyda-Legutko 1987: 54, 139; cf. Schmiedehelm 2011: 144 here as Czasza). 3. Janowo (Jędrychowo-J.), powiat Mrągowo (former Heinrichsdorf, Kreis Sensburg), Poland, grave 341. Type AG36. H. 77 mm. Unpublished (PM-A 1610/ bronze rivets are mentioned here; cf. Schmiedehelm, personal files: 7.20d 9; Schmiedehelm 2011: 144). 4. Stryczowice, powiat Ostrowiec Świętokrzyski, Poland, grave. Type AG36. H. 108 mm (fig. 5. 7) (Liana 1962: 49, pl. XIV. 2, 6; Madyda-Legutko 1987: 168 here as type AG35 45; 1990: 572, 583, fig. 8. 3). 5. Drozdowo, powiat Łomża, Poland, grave 37. Type AG42 (perhaps with bronze rivets). H. 99 mm. Unpublished (Muzeum Północno-Mazowieckie in Łomża, personal communication from Mieczysław Bienia). List 3. Iron belt buckles type AG35 45 decorated with silver inlays 1. Badow, Landkreis Nordwestmecklenburg (Mecklenburg-Vorpommern), Germany, grave 658. Type AG37 (?). H. 31 mm (Bemmann 1999: 101, pl. 100/658. 2). 2. Hagenow, Kreis Ludwigslust (Mecklenburg-Vorpommern), Germany, grave VII/1907. Type AG43/44. H. ca. 44 mm (Geisler 1975: 192, pl. 26. a c; Madyda 1990: 560, fig. 3. 4; Voß 2008: 263, fig. 5. 1). 3. Hagenow, Kreis Ludwigslust (Mecklenburg-Vorpommern), Germany, grave 5/1995. Type AG37 (frame not preserved, perhaps with silver inlays?). H. ca. 38 mm (Voß 2008: 256, 263, fig. 5. 2). 4. Oblin, powiat Garwolin, Poland, grave 46b. Type AG37. H. 74 mm (Czarnecka 1992: 42, pl. IV. c; 2007: 22, pl. XLIX. 45b. 18 here as type AG37 39). 5. Starachowice, powiat Starachowice, Poland, grave (?). Unique type with uni-partite frame and internal bar with two tongues, perhaps with silver inlays. H. ca. 57 mm. Unpublished (Roel, personal files). Bibliography Archival sources Grenz, personal files Personal files of Rudolf Grenz, Archäologisches Landesmuseum, Stiftung Schleswig- Holsteinische Landesmuseen Schloß Gottorf, Schleswig. Jahn, personal files Personal files of Martin Jahn, Instytut Archeologii Uniwersytetu Warszawskiego. Jankuhn, personal files Personal files of Herbert Jankuhn, Archäologisches Landesmuseum, Stiftung Schleswig- Holsteinische Landesmuseen Schloß Gottorf, Schleswig. PM Inventarbücher Inventory books of former Prussia-Museum in Königsberg, now Kaliningradskiy oblastnoy istoriko-khudozhestvennyy musey. PM-A Fundarchiv des Prussia-Museum (Fundakten), Museum für Vor- und Frühgeschichte Staatliche Museen zu Berlin. Roel, personal files Personal files of Stefan Roel, Państwowe Muzeum Archeologiczne, Warszawa. Schmiedehelm, personal files Personal files of Martha Schmiedehelm, Ajaloo Instituut, Tallin (courtesy of Anna Juga-Szymańska, Instytut Archeologii Uniwersytetu Warszawskiego). Secondary literature Bemmann G. Badow. Ein Gräberfeld der jüngeren vorrömischen Eisenzeit und älteren römischen Kaiserzeit im Landkreis Nordwestmecklenburg // Beiträge zur Ur- und Frühgeschichte Mecklenburg-Vorpommerns. 34. Lübstorf, Bitner-Wróblewska A., Nowakiewicz T., Rzeszotarska-Nowakiewicz A., Wróblewski W. Treść i znaczenie odzyskanych ksiąg inwentarzowych Prussia-Museum // Archeologiczne księgi inwentarzowe dawnego Prussia- Museum / Aestiorum Hereditas. I. Olsztyn, Bitner-Wróblewska A., Nowakiewicz T., Rzeszotarska-Nowakiewicz A. Katalog // Archeologiczne dziedzictwo Prus Wschodnich w archiwum Feliksa Jakobsona / Aestiorum Hereditas. II. Warszawa, 2011.

21 Bronze Belt Buckles with Doubled Tongue between Scandinavia and the Black Sea. In Search of Local and Inter-Regional Connections during the Roman Period 21 Bierbrauer V. Die Goten vom Jahrhundert n. Chr.: Siedelgebiete und Wanderbewegungen aufgrund archäologischer Quellen // Peregrinatio Gothica. III / Universitetets Oldsaksamlings Skrifter, Ny rekke. 14. Oslo, Blankenfeldt R. Studien zu den Persönlichen Ausrüstungen aus dem Thorsberger Moor (typescript of PhD thesis). Kiel, Carnap-Bornheim C. von. Zu den Prachtgürteln aus Neudorf-Bornstein (Kreis Rendsburg-Eckernförde) Grab 3 und Grab 7 // Drik og du vil leve skønt. Festskrift til Ulla Lund Hansen på 60-årsdagen 18. august 2002 / Publications from the National Museum. Studies in Archaeology & History. 7. Copenhagen, Carnap-Bornheim C. von. The ornamental belts from Ejsbøl Bog and Neudorf-Bornstein // The spoils of Victo ry The North in the shadow of the Roman Empire. Copenhagen, Ciesliński A. Kulturelle Veränderungen und Besiedlungsabläufe im Gebiet der Wielbark-Kultur an Łyna, Pasłęka, und oberer Drwęca // Berliner Beiträge zur Vor- und Frühgeschichte. NF. 17. Berlin, Czarnecka K. Two newly-found Roman swords from the Przeworsk culture cemetery in Oblin, Siedlce District, Po land // Journal of Roman Military Equipment Studies Czarnecka K. Oblin. Ein Gräberfeld der Przeworsk-Kultur in Südmasowien // Monumenta Archeologica Barbarica. XIII. Warszawa, Dąbrowska T. Kamieńczyk. Ein Gräberfeld der Przeworsk-Kultur in Ostmasowien // Monumenta Archaeologica Barbarica. III. Kraków, Dinnyés I. A hévizgyörki szarmata sírok // Studia Comitatensia Eggers H. J., Stary P. F. Funde der Vorrömischen Eisenzeit, der Römischen Kaiserzeit und der Völkerwanderungszeit in Pommern // Beiträge zur Ur- und Frühgeschichte Mecklenburg-Vorpommerns. 38. Lübstorf, Engel C. Das jüngste heidnische Zeitalter in Masuren // Prussia. Zeitschrift für Heimatkunde und Heimatschutz Engelhardt C. Horsbjerg Mosefund. Beskrivelse af de Oldsager, som i Aarene ere Udgravede af Thorsbjerg Mose ved Sønder-Brarup i Angel // Sønderjydske Mosefund. I. Kjöbenhavn, Garbsch J. Die norisch-pannonische Frauentracht im 1. und 2. Jahrhundert. München, Geisler H. Das germanische Urnengräberfeld bei Kemnitz, Kreis Potsdam-Land. I. Katalog // Veröffentlichungen des Museums für Ur- und Frühgeschichte Potsdam. 8. Berlin, Geisler H. Eine Doppeldornschnalle der frühen römischen Kaiserzeit aus Hagenow // Ausgrabungen und Funde /4. Geisler H. Das germanische Urnengräberfeld bei Kemnitz, Kreis Potsdam-Land. II. Text // Veröffentlichungen des Museums für Ur- und Frühgeschichte Potsdam. 18. Berlin, Gładykowska-Rzeczycka J. J., Sokół A., Pudło A., Ziętek A. Stan zdrowotny mieszkańców Pruszcza Gdańskiego z okresu rzymskiego, pochowanych na cmentarzysku stanowisko 7 // XIII Sesja Pomorzoznawcza. 1. Od epoki kamienia do okresu rzymskiego. Gdańsk, Godłowski K. Die Chronologie der jüngeren und späten Kaiserzeit in den Gebieten südlich der Sudeten und Karpaten // Probleme der Relativen und Absoluten Chronologie ab Latènezeit bis zum Frühmittelalter. Materialien des III. Internationalen Symposiums: Grundprobleme der frühgeschichtlichen Entwicklung im nördlichen Mitteldonaugebiet, Kraków-Karniowice Dezember Kraków, 1992a. Godłowski K. Germanische Wanderungen im 3. Jh. v. Chr. 6. Jh. n. Chr. und ihre Wiederspiegelung in den historischen und archäologischen Quellen // Peregrinatio Gothica. III / Universitetets Oldsaksamlings Skrifter, Ny rek ke, 14. Oslo, 1992b. Godłowski K., Wichman T. Chmielów Piaskowy. Ein Gräberfeld der Przeworsk-Kultur im Świętokrzyskie-Ge birge // Monumenta Archaeologica Barbarica. VI. Kraków, Heydeck J. Das Gräberfeld zu Korklack, Kreis Gerdauen // Sitzungsberichte der Altertumsgesellschaft Prussia zu Königsberg in Pr ( ). Janikowski J. Międzyrzecze Wisły, Drwęcy i Osy w okresach: późnolateńskim i wpływów rzymskich (typescript of PhD thesis). Toruń, Juga A., Ots M., Szymański P. Über die Vorteile der Bildung einer didaktischen Kolektion. Materialien der Bogaczewo-Kultur und Olsztyn-Gruppe in Ajaloo Instituut in Tallinn (Estland) // Antyk i barbarzyńcy. Księga dedykowana Profesorowi Jerzemu Kolendo w siedemdziesiątą rocznicę urodzin. Warszawa, Kaszewska E. Witaszewice, distr. de Łęczyca, dép. de Łódź // Inventaria Archaeologica. Pologne XXVII. Kleemann O. Die Funde des Elbinger Kreises im Prussia-Museum // Festschrift für Bruno Ehrlich zum 70. Geburtstag dargebracht = Elbinger Jahrbuch XV.

22 22 Jacek Andrzejowski, Renata Madyda-Legutko Kostrzewski B. Cmentarzyska z okresu rzymskiego w Czaczu i w Kokorzynie w pow. kościańskim i w Pięczkowie w pow. średzkim // Fontes Archaeologici Posnanienses VI (1955). Kramer S. Die Grabung auf dem kaiserzeitlichen Gräberfeld Kemnitz, Kreis Potsdam-Land // Ausgrabungen und Funde / 6. Kunkel O. Urgeschichte, Volkskunde, Landesgeschichte und Stadtkultur, kirchliche Kunst. (Mitteilungen aus dem Pommerschen Landesmuseum.) Urgeschichte // Baltische Studien Liana T. Znaleziska z okresu późnolateńskiego i rzymskiego z terenu województwa kieleckiego w zbiorach Państwowego Muzeum Archeologicznego // Wiadomości Archeologiczne XXVIII / 1. Lissauer A. Die Prähistorischen Denkmäler der Provinz Westpreussen und der angrenzenden Gebiete. Leipzig, Madyda-Legutko R. Die Gürtelschnallen der Römischen Kaiserzeit und der frühen Völkerwanderungszeit im mitteleuropäischen Barbaricum // BAR IS. 360 (1986). Oxford, Madyda-Legutko R. Doppeldornschnallen mit rechteckigem Rahmen im europäischen Barbaricum // Germania / 2. Madyda-Legutko R. Uwagi o ornamentyce ażurowej w kręgu zachodniobałtyjskim // Bałtowie i ich sąsiedzi. Marian Kaczyński in memoriam / Seminarium Bałtyjskie. II. Warszawa, Madyda-Legutko R. Studia nad zróżnicowaniem metalowych części pasów w kulturze przeworskiej. Okucia końca pasa. Kraków, Madyda-Legutko R., Rodzińska-Nowak J., Zagórska-Telega J. Wyniki dalszych badań na cmentarzysku kultury przeworskiej w Prusieku stan. 25, pow. Sanok // Rocznik Przemyski (Archeologia) XLIII / 2. Madyda-Legutko R., Rodzińska-Nowak J., Zagórska-Telega J. Opatów, Fpl. 1. Ein Gräberfeld der Przeworsk- Kultur im nordwestlichen Kleinpolen. Katalog // Monumenta Archaeologica Barbarica. XV/1. Warszawa; Kraków, 2011a. Madyda-Legutko R., Rodzińska-Nowak J., Zagórska-Telega J. Opatów, Fpl. 1. Ein Gräberfeld der Przeworsk-Kultur im nordwestlichen Kleinpolen. Tafeln // Monumenta Archaeologica Barbarica. XV/1. Warszawa; Kraków, 2011b. Martin F. R. Fibulor och soljor från Kertsch // KVHAA Månadsbladet. 32. Stockholm, Mączyńska M. Bemerkungen über einige Typen der Augenfibeln im Barbaricum // Zentrum und Peripherie Gesellschaftliche Phänomene in der Frühgeschichte, Materialien des 13. Internationalen Symposiums Grundprobleme der frühgeschichtlichen Entwicklung im mittleren Donauraum Zwettl, Dezember 2000 / Mitteilungen der Prähistorischen Kommission. 57. Wien, Motyková Šneidrová K. Zur Chronologie der ältesten römischen Kaiserzeit in Böhmen // Berliner Jahrbuch für Vorund Frühgeschichte Natuniewicz-Sekuła M., Okulicz-Kozaryn J. Weklice. A Cemetery of the Wielbark Culture on the Eastern Margin of Vistula Delta (Excavations ) // Monumenta Archaeologica Barbarica. XVII. Warszawa, Nowakowski W. Kulturowy krąg zachodniobałtyjski w okresie wpływów rzymskich. Kwestia definicji i podziałów wewnętrznych // Archeologia bałtyjska. Materiały z konferencji, Olsztyn kwietnia 1988 roku. Olsztyn, Nowakowski W. Kultura przeworska a zachodniobałtyjski krąg kulturowy // Kultura Przeworska. I. Lublin, Nowakowski W. Kultura wielbarska na wschód od dolnej Wisły. Materiały z dawnych badań i przypadkowych odkryć w zbiorach berlińskich i norymberskich // Barbaricum. 3. Warszawa, Nowakowski W. Das Samland in der römischen Kaiserzeit und seine Verbindungen mit dem römischen Reich und der barbarischen Welt // Veröffentlichungen des Vorgeschichtlichen Seminars Marburg. 10. Marburg; Warszawa, Nowakowski Z. Cmentarzysko kultury przeworskiej w Żdżarowie, pow. sochaczewski // Wiadomości Archeologiczne LVI ( ). Raddatz K. Der Thorsberger Moorfund. Gürtelteile und Körperschmuck // Offa-Bücher. 13. Neumünster, Raddatz K. Der Wolka-See, ein Opferplatz der Römischen Kaiserzeit in Ostpreußen // Offa /50 ( ). Rewekant A., Segieda S. Wyniki badań antropologicznych szczątków kostnych ludności kultury wielbarskiej pochowanej na cmentarzysku w Kowalewku, gm. Oborniki // Archeologiczne badania ratownicze wzdłuż trasy gazociągu tranzytowego. II. Wielkopolska. 3. Poznań, Rasmussen B. M. Brokær. Ein Reichzentrum der römischen Kaiserzeit in Südwestjütland // Acta Archaeologica

23 Bronze Belt Buckles with Doubled Tongue between Scandinavia and the Black Sea. In Search of Local and Inter-Regional Connections during the Roman Period 23 Romanowska-Grabowska O. Badania ratownicze na cmentarzysku z okresu rzymskiego w Ruży (Róży), pow. Golub-Dobrzyń w roku 1968 // Komunikaty Archeologiczne. Badania wykopaliskowe na terenie województwa bydgoskiego w latach Bydgoszcz, Romanowska-Grabowska O., Janikowski J. Wyniki badań na cmentarzysku ciałopalnym z okresu wpływów rzymskich w Rużu (Różu), dawny pow. Golub-Dobrzyń // Rocznik Muzeum w Toruniu VIII. Schmiedehelm M. Das Gräberfeld am Jaskowska-See in Masuren. Studien zur westmasurischen Kultur der römischen Eisenzeit. Warszawa, Skorupka T. Kowalewko 12. Cmentarzysko birytualne ludności kultury wielbarskiej (od połowy I w. n. e. do początku III w. n. e.) // Archeologiczne badania ratownicze wzdłuż trasy gazociągu tranzytowego. II. Wielkopolska. 3. Poznań, Sobczyk P. Penetracje terenowe Działu Archeologicznego MZKiD w latach // Rocznik Muzealny (Włocławek) X. Sobczyk P. Wielokulturowe stanowisko nr 2 w Kołacie na Ziemi Dobrzyńskiej // XVII Sesja Pomorzoznawcza. 1. Od epoki kamienia do wczesnego średniowiecza. Gdańsk, Szela A. Badania i pytania. Problemy związane z Mazowszem Północnym w pierwszych wiekach naszej ery // Blisko i daleko. Księga jubileuszowa Instytutu Archeologii Uniwersytetu Warszawskiego. Warszawa, Szter I. Cmentarzysko z okresu wpływów rzymskich i z okresu wędrówek ludów w Kamieniu na Pojezierzu Mazurskim // Wiadomości Archeologiczne LXI ( ). Vaday A. H. Der Grabfund von Gyulavári // Communicationes Archaeologicae Hungariae. Budapest, Vasil yev A. Druzhnoye and Thorsberg: some aspects of the study of belt sets from the Late Roman Period // Inter Ambo Maria. Contacts between Scandinavia and the Crimea in the Roman Period. Kristiansand; Simferopol, Voß H.-U. Zwischen Vannius-Reich und Vimose Die elitären Krieger von Hagenow // Aktuelle Forschungen zu Kriegsbeuteopfern und Fürstengräbern im Barbaricum. Internationales Kolloquium, Schleswig Juni 2006 / Schriften des Archäologischen Landesmusems. Ergänzungsreihe. 4. Neumünster, Walenta K. Leśno i mikroregion w okresie rzymskim. Chojnice, Wichrowski Z. Kraśnik Piaski, pow. kraśnicki, woj. lubelskie // Wandalowie strażnicy bursztynowego szlaku. Państwowe Muzeum Archeologiczne w Warszawie, 8 marca 16 czerwca Katalog wystawy. Lublin; Warszawa, Wolfram H. Die Goten. Von den Anfängen bis zur Mitte des sechsten Jahrhunderts. Entwurf einer historischen Ethnographie. München, Zielonka B. Stanowisko wielokulturowe w Adolfinie w pow. aleksandrowskim // Przegląd Archeologiczny XIII. Zimmer-Linnfeld K. Westerwanna I // Beiheft zum Atlas der Urgeschichte. 9. Hamburg, Бажан И. А., Герцегер Д. С. Бронзовая пряжка римского времени из Ольвии // Петербургский археологический вестник Васильев А. А. Германские воинские пряжки и наконечники ремней на Боспоре // Боспорские исследования. IX. Симферополь; Керчь, Гудкова А. Б., Фокеев М. М. Земледельцы и кочевники в низовьях Дуная I IV вв. н. э. Киев, Кулаков В. И. Доллькейм-Коврово. Исследования 1879 г. // Prussia Antiqua. 2 (Monumenta). Минск, Радюш О. А., Скворцов К. Н. Находки деталей щитов в ареале самбийско-натангийской культуры // Ger mania Sarmatia: Древности Центральной и Восточной Европы эпохи римского влияния и переселения народов. Калининград, Русанова И. П. Вельбарская культура // Славяне и их соседи в конце I тысячелетия до н. э. первой половине I тысячелетия н. э. Археология СССР. М., Сегеда С. П., Дячeнко В. Д. Антропологический материал из черняховского могильника у с. Холмское // Петербургский археологический вестник Шаров О. В. Данные письменных и археологических источников о появлении германцев на Боспоре (проблема выделения «германских древностей» на Боспоре) // Stratum plus Щукин М. Б. О первом появлении готов в Дунайско-Причерноморском регионе и начале черняховской культуры // Европа-Азия: проблемы этнокультурных контактов. K 300-летию Санкт-Петербургa. СПб., Щукин М. Б. Готский путь. Готы, Рим и черняховская культура. СПб., 2005.

24 24 Jacek Andrzejowski, Renata Madyda-Legutko Яцек АНДЖЕЁВСКИЙ, Рената МАДЫДА-ЛЕГУТКО Бронзовые поясные пряжки с двойным язычком между Скандинавией и Причерноморьем. В поисках местных и межрегиональных связей в римское время Резюме Широкий пояс с массивной прямоугольной пряжкой с двойным язычком был характерным элементом воинского костюма на территории центрально-европейского Барбарикума в фа зе B2b раннеримского времени и в начале позднеримского времени. Эти пряжки были подробно рассмотрены Р. Мадыдой-Легутко (Madyda-Legutko 1987; 1990), которая отнесла их к типам группы G. Однако если Р. Мадыде- Легутко удалось зафиксировать 105 экземпляров пряжек с двойным язычком, в настоящее время их количество составляет почти 250 экземпляров, причём немногим больше 30 штук, то есть чуть более 13 %, сделаны с применением бронзы (см. Список 1). Зона распространения бронзовых пря моугольных поясных пряжек с двойным язычком определённо отличается от области распространения железных изделий (рис. 1). В частности, на территории вельбарской культуры встречаются почти исключительно брон зовые пряжки здесь зафиксировано не более трёх железных изделий. Бронзовые пряж ки редко находят на землях западных бал тов (здесь их всего пять из более чем 110 экземпляров) и на территории пшеворской культуры (всего четыре из 80 экземпляров), а к югу от Карпат прямоугольных поясных пряжек с двойным язычком нет вообще, если не считать обнаруженных здесь несколько железных вещей, тогда как из Северного Причерноморья происходят почти исключительно бронзовые находки. В эльбской зоне бронзовые поясные пряжки также встречаются редко, зато они составляют значительный процент находок из Ютландии и с датских островов (пять из 14 экземпляров). По археологическим данным, бронзовые, равно как и железные, поясные пряжки с двойным язычком характерны для мужского убора (Асдруп на территории Дании, Прусек и могила 1876 г. в Подвеске на территории Польши, Совхозное в России, Спыхувко в Польше, Ольвия на Украине, Виташевице в Польше, а также, вероятно, Торсберг в Дании и, быть может, Керчь на Украине). Мужской пол погребённых в некоторых других могилах (в Холмском на Украине, Ковалевко, Прусеке и Прущ-Гданьском в Польше) определён на основании только антропологических исследований. Эта закономерность не слишком строгая, поскольку в некоторых случаях, судя по погребальному инвентарю, элементами женского убора могут быть и крупные бронзовые поясные пряжки, соответствующие железным пряжкам от так называемых портупей для мечей (Руже, Елонки, могила 448 в Веклице, Славоше и, вероятно, Лесьна все на территории Польши). Одна бронзовая поясная пряжка с двойным язычком происходит из погребения ребёнка, сопровождавшегося определённо женским набором инвентаря (Кемниц на территории Германии). Большинство бронзовых поясных пряжек с двойным язычком надёжно датированы периодом B2/C1 C1a, что соответствует хронологии железных находок. Исключениями являются пряжки из Кемница, могилы 139 в Веклице и особенно могилы 149В в Прущ- Гданськом 7. Последнее погребение по ин вентарю датируется периодом В2а (рис. 2), то есть эта пряжка представляет собой древнейшую из известных находок прямоугольных поясных пряжек с двойным язычком на территории Барбарикума. Бронзовые поясные пряжки с двойным язычком образуют группу, довольно раз нообразную как по конструкции, так и по стилю. Выраженный индивидуальный характер отдельных экземпляров, вероятно, является

25 Bronze Belt Buckles with Doubled Tongue between Scandinavia and the Black Sea. In Search of Local and Inter-Regional Connections during the Roman Period 25 результатом свойств материала, позволявших моделировать законченное изделие гораздо свободнее, чем при изготовлении вещей из железа. Среди бронзовых поясных пряжек с двойным язычком преобладают экземпляры с выраженно прямоугольными рамками, высо та которых в два-три раза больше ши рины, двухчастные, обычно с узким пря моугольным щитком. Среди них чаще всего встречаются находки с вилкообразным языч ком, напоминающие тип AG37 (рис ; ). Реже встречаются бронзовые двухчастные поясные пряжки с Н-образным язычком, имеющие два выступа на оси и соответствующие типу AG43 (рис ). Вероятно, их вариантом является поясная пряжка со слегка скруглённой рамкой и двумя подвижными язычками, которые фиксируются щитком, из Лазовского в Самбии (Россия) (рис. 5. 5). Экземпляры с одночастной рамкой встречаются редко две или три такие пряжки соответствуют типу AG36 (рис ), а две другие относятся соответственно к типам AG35/36 и AG42 (рис ). Наконец, примерно четыре бронзовые поясные пряжки с более широким основанием двойного язычка в целом напоминают тип AG45 (рис ). Значительная распространённость широких поясов с прямоугольными пряжками с двумя язычками и массивными наконечниками, характерных для воинского убора, в центральноевропейском Барбарикуме в фазах B2b C1a (рис. 1), вероятно, доказывает, что разные группы воинов в разной культурной среде поддерживали тесные контакты друг с другом. В частности, это касается воинов пшеворской, западнобалтских богачевской и самбийсконатангийской (Доллькайм-Коврово), а также, вероятно, вельбарской культур. С другой стороны, распространение таких, сделанных из бронзы, поясных пряжек заставляет предположить, что центром их производства и использования была вельбарская культура. Вероятно, большая часть бронзовых поясных пряжек с двойным язычком, происходящих с территории распространения железных находок в пшеворской культуре, а также из зоны западных балтов, сделана в вельбарских мастерских. С этих территорий схема поясной пряжки с двойным язычком могла распространиться вплоть до эльбской зоны, Ютландии и датских островов, где её, вероятно, воспроизводили в местных мастерских (Саддеруп, Торнхольм и Асдруп в Дании). Однако некоторые другие бронзовые пряжки (из Кемница, Раушендорфа и Торсберга на территоии Германии), вероятнее всего, были привезены с вельбарской или западнобалтской территории. На территорию к югу от Карпат поясные пряжки с двойным язычком, вероятно, привозили из ареала пшеворской культуры, тогда как исключительно бронзовые поясные пряжки Севрного Причерноморья, вероятнее всего, следует связывать с проникновением в понтийскую зону носителей вельбарской культуры, которая в фазе B2/C1 C1a распространилась на обширную территорию, прежде бывшую восточной зоной пшеворской культуры, достигнув Волыни. Незаконченный вилкообразный язычок из с. Девочки на Украине (рис. 3. 3) связывает поясные пряжки Северного Причерноморья с зоной поселений вельбарской культуры.

26 26 Ruth BLANKENFELDT Weapon Deposits in the Thorsberger Moor 150 Years of Research and New Рerspectives The Thorsberger Moor: introduction It was through reports of conspicuous objects found during cutting of peat in the Thorsberger Moor (Süderbrarup, district Schleswig-Flensburg) (fig. 1), that the curator of the Sammlung nordischer Altertümer, Helvig Conrad Engelhardt, became aware of the location. After numerous viewings of the site, he conducted excavations during the summer months between 1858 and 1861 probably unaware of the fact that he would lay the foundation for a new archaeological genre! In total, he laid free through 750 m 2 of the relatively dry bog under technically difficult conditions and discovered numerous artefacts that showed a strong military component (Engelhardt 1863) (fig. 2). During the following years, Engelhardt carried out further excavations, including locations in different Danish bogs. At the sites at Nydam and Porskjær in Jutland, as at the sites of Vimose and Kragehul on Funen, he noticed similarities in the composition of the archaeological inventory to those discovered in the Thorsberger Moor (Engelhardt 1865; 1866; 1867; 1869) (fig. 1): There was also a great number of items found in a comparatively small area in these locations. In addition, most of the finds like shields, swords, lances and spears showed a high percentage of objects that could be described as militaria. Due to the quickly publication with high quality find-drawings these locations became accessible to a wider audience. The unusual composition of the inventory rapidly led to a variety of interpretations. Among other things, they were declared as battlefield relics, as remains of warriors who had broken through the ice, as fune rary offerings for deceased warriors in Valhalla, or as abandoned large-scale deposits of metal merchants (Lund Hansen 2003; Rau, Carnap-Bornheim 2012: 517 f.). The Danish archaeologist Jens Jacob Asmussen Worsaae initially declared the artefacts to be offerings consisting of weapons that were captured as war trophies, without having concrete evidence to reach this conclusion (Worsaee 1865: 55 f.). Engelhardt was more cautious in his interpretation of the findings, however he assumed them to be objects that were intentionally deposited. This included the often deliberate sorting of artefacts according to groups of objects, and the different elevation levels of various objects, indicating a difference in time line (Engelhardt 1866: 23 f.; Wiell 2003). 1 Contemporary interpretation of the weapon deposits More than 150 years have passed since the first archaeologically motivated prick of a spade in the Thorsberger Moor. Meanwhile, over two dozen similar sites are known. Geographically, the offering places stretch from Schleswig Holstein in northern Germany via mainland Denmark (Jutland) and the Danish islands to mainland Sweden and the Swedish island of Öland (fig. 2). 2 They encompass a chronological frame of the Middle Pre-Roman Iron Age to the end of the Migration period (summarized: Ilkjær 1990a: 332, fig. 201; 2003a: 46, fig. 2; 2003b). The decisive breakthrough in reaching a certain interpretation of religiously motivated depositions of the armour of vanquished opponents 1 Numerous references to Engelhardt s observations during the excavations are also found in the excavation diary, dated 1860, which was rediscovered in the inventory of the National Museum of Copenhagen in For the history of research and interpretation of these votive sites, refer to: (Blankenfeldt, Rau 2009; Ilkjær, Iversen 2009; Carnap-Bornheim, Rau 2012; Rau 2012; Rau, Carnap-Bornheim 2012).

27 Weapon Deposits in the Thorsberger Moor 150 Years of Research and New Рerspectives 27 Fig. 1. Places with sacrifices of military equipment (Blankenfeldt, Rau 2009) was reached in the second half of the twentieth century. The modern excavated and documented sites Ejsbøl (Ørsnes 1970; 1984; 1988; Andersen 2003) and Illerup Ådal 3 in Jutland as well as follow-up inspections of Nydam, which had previously been examined by Engelhardt (Bemmann, Bemmann 1998a; 1998b; Rau 2010), played a central role. Following his excavations in Ejsbøl, M. Ørsnes was the first to publish a catalogue of matching objects that had been forcibly broken prior to being sacrificed and that were discovered far apart from each other (Ørsnes 1970; 1988: fig. 43). With this so called fragment matchings, he was able to demonstrate that a great number of votive offerings found in a large area came to rest at the exact same time. This is even more clearly evident for the findings from Illerup Place A. Since here the sacks and bundles of weapons and armour that were sunk by boat could be brought into relation to one another (see, for example, Ilkjær 1993a: 16 f., fig. 1; 2002: 28 f., with text illustration). With approximately 15,000 objects, the votive site Illerup Place A, dating to the early third century AD, could be classified as the largest closed find dating from Scandinavian prehistory (Ilkjær 1990a: 24 f., fig. 14, 15). Due to the good preservation conditions in the sacrificial places, the large amount of findings include not only metal objects, but also objects consisting of wood, leather, textile, bone, antler, horn or plant fibres, and thus artefacts that have rarely survived in graves and settlements. 3 In addition to numerous articles on the site, the individual groups of objects are being processed in monographic form and currently appear (Ilkjær 1990a; 1990b; 1993a; 1993b; 2001a; 2001b; Carnap-Bornheim, Ilkjær 1996a; 1996b; 1996c; 1996d; Bursche 2011; Biborski, Ilkjær 2006a; 2006b; Pauli Jensen, Nørbach 2009). Summary to the discovery site Illerup Ådal: (Ilkjær 2002).

28 28 Ruth BLANKENFELDT Fig. 2. Finds from the Thorsberger Moor (photo: Regional Museum of Archaeology Schleswig)

29 Weapon Deposits in the Thorsberger Moor 150 Years of Research and New Рerspectives Modern topographical, geological and palynological studies have also shown that the widely used term bog sacrifices is incorrect in most cases. Thus, the larger deposits from the third and fourth century AD, in Illerup Ådal, Vimose, Ejsbøl and Nydam, were undertaken in open seas, or at least in water-filled basins. Evidence proving votive offerings on a surface accessible by foot on the now peat-filled sea, in Nydam for example, only appear in the fifth century AD. The deposit of military equipment represents a prominent archaeological source group, often referred to as spoils of war offering sites / Kriegsbeuteopferplätze in the literature (see e. g. Sieg 2003; Aktuelle 2008). The artefacts themselves are interpreted as being captured military equipment from conquered army units, which were taken from those defeated in inner-germanic conflicts. These inferior opponents did not originate from the geographical area in which the objects were sacrificed. Therefore the artefacts are interpreted as non-local objects (Ilkjær 1993a: 374 f.). Traces of destruction on the objects bear witness to a preceding battle (see e. g. Blankenfeldt, Rau 2009: 138, fig. 8). In addition, specific patterns on, but also the systematic destruction of, some artefacts prior to the deposit prove ritual treatment thereof (most recent: Lau 2009). Therefore, the procedures of deposit are certainly to be regarded as religiously motivated the addressee, however, remains unknown. Ancient written sources which could provide hints about these ceremonies are completely missing. However, there are descriptions of votive practices of other barbaric groups. For example the report of Paulus Orosius from the year 105 BC, about the victorious Cimbri and Teutoni: The enemies seized both encampments and immense spoils and, in a new and unusual curse, destroyed everything they had captured: clothing was torn and thrown away, gold and silver were thrown into the river, the armour of the men was destroyed, the (face and chest) adornments of the horses were ruined, and the horses themselves drowned in the river; nooses were placed around the necks of the people and they were hung from the trees, so that the victors knew nothing of the spoils, and the defeated knew nothing of mercy (Paul. Oros. Hist. adv. pagan ; cited and translated by Altes 1995: 219). 29 Chronological studies utilizing the finds from the sites with weapon deposits Essential to the chronological key of the various deposits within a find site, gave insight that contemporaneous lance- and spearheads, and also shield components, are very similar in form over a large geographical area. Moreover these subjects performed a synchronous, clearly describable, structural change (Ilkjær 1975; 1990a; 1990b; 2003b: 29 f., fig 21; see also Bemmann, Hahne 1994) (fig. 3). Since the fundamental chronological studies on the period of the Roman Iron Age by H.-J. Eggers, the designation of the stages B1 B2 C1 C2 C3 has been retained, but has gone through numerous modifications (Eggers 1951; 1955). Thereby, the period of the Marcomannic Wars (AD ) is an important marker for the transition from the Older to the Younger Roman Iron Age. The creation of a trans-spatial system of chronology is still difficult and the trans-regional synchronization often unsuccessful due to cultural traits. 4 The study of the votive sites and their military equipment with a large amount of simultaneously deposited artefacts, led to a large variety of new and detailed chronological statements concerning the Roman Iron Age and the Migration period. These studies are strongly coupled to specific groups of weaponry objects. It was Jorgen Ilkjær who laid the foundation for a descriptive system with fixed morphological characteristics that can be used as chronological indicators (Ilkjær 1990a: 29 f.). Through the analysis of a bundle of weapons from the sacrificial site of Vimose, which contained lance- and spearheads that had been simultaneously deposited during the Young Roman Iron Age, stage C1b, he created the basis for the grouping of individual types with solid morphological characteristics (Ilkjær 1975; 1990a: 29 f., 40, tab. 16). This system, which is developed independently of coins, fibulae and imported objects, uses the marked homogeneity of lance- and spearheads crafted in the same time frame in the Younger Roman Iron Age and the Migration period. Likewise the equality indicates a large-scale production in central workshops (Jørgensen 2001). For dating references of artefacts from the northern European Barbaricum, studies based on 4 A table with various chronological systems for this time period is pictured at: (Lund Hansen 1987: 30, fig. 10).

30 30 Ruth BLANKENFELDT combined groups of weapons from graves and weapon deposits are of great importance. Besides the combined groups 1 12 with chronological relevance worked out by Ilkjær (1990a), J. Bemmann and G. Hahne also authored a similar dating system based on the inventories from Norwegian weapon graves (Bemmann, Hahne 1990). Meanwhile, these systems could be modified further. In this context, especially the sacrifice sites of Nydam (Rau 2010) and Kragehul (Iversen 2008a; 2008b; 2010) are to be mentioned. By understanding the modern excavated sites and chronology of weapons, the old excavated finding-places can be reinterpreted today as well. All in all in sacrificial sites with offerings of military equipment, few and chronologically differentiated extensive deposits were performed. While H. Jankuhn assumed in his investigations into the Thorsberger Moor that sacrifices were offered at regular intervals (Jankuhn 1936a; 1936b), this is meanwhile clearly refuted. Thus, analysis of the finds from Illerup Ådal for example, concluded that a time span of up to 100 years could pass between deposits (in summary: Ilkjær 1990a: 332, fig. 201; 2003a: 46, fig. 2; 2003b). Thorsberger Moor: the site The composition and quantity of the find ensembles from the votive sites is influenced by various selective factors: from the looting to the sacrifice, to the excavation methods and the archiving. In addition morphological conditions of the respective sites impact the information about the sacrificed objects as well. The physicochemical conditions of the Thorsberger Moor with a very low PH balance have the effect that only artefacts consisting of coloured or precious metals, wood, leather, glass or ceramics survived. Objects made of iron, antler or bone (apatite) have not been preserved. Today, the majority of these artefacts are located in the Archaeological National Museum, Foundation Schleswig-Holstein State Museums Gottorf Castle, Schleswig, and in the National Museum in Copenhagen. Following publication by the excavator (Engelhardt 1863), the finds were discussed in particular by K. Raddatz (1952; 1957; 1957/58; 1987a; 1987b), as well as in subsequent publications (Werner 1941; Carnap-Bornheim 1997; Neben 2004; Blankenfeldt 2008). The arte- Fig. 3. The possible chronological frame and origin of the warriors represented in the weapon deposits in Thorsberger Moor

31 Weapon Deposits in the Thorsberger Moor 150 Years of Research and New Рerspectives facts were recently catalogued and adapted to the current state of research in a multi-year international project. 5 Fundamental questions concerning offering places with sacrifices of military equipment places the focus on the time frame and the number of deposits. Since the objects from these localities are non-local spoils, it is also necessary to clarify from which geographical and cultural contexts the defeated warriors originated. Due to the specific morphological conditions, significant find groups were not preserved in the Thorsberger Moor. This primarily includes chronological dial indicators in form of lance- and spear-heads. In addition, no objects consisting of horn and antler have been preserved. Therefore, objects like combs, for example, are missing, which, due to raw material and shape analysis, would be an important foundation for the interpretation of the origin of the warriors (Ilkjær 1993a: 271 f., 374 f.; 2003: 38 f.). Questions about the number and time frame of sacrifices in the Thorsberger Moor, as well as to the origin of the warriors represented, have been discussed based primarily upon the brooches (Ilkjær, Lønstrup 1982; Lønstrup 1984). According to these, three deposits of military equipment were performed here, the most extensive taking place between AD (Ilkjær 1990a: 290, 324 f.; fig. 201; 2003a: 46, fig. 2). Because of comparable fibula forms the origin of the warriors from this deposit has been described with the northern Elbe and the Rhine-Weser regions (Ilkjær, Lønstrup 1982: 96 f.; Lønstrup 1984: 97 f.). Two further deposits were significantly less extensive (Lønstrup 1984: 92, fig. 1 2; Ilkjær 1990a: 284, 301, 324 f.; fig. 201; 2003a: 46, fig. 2). Consequently, the first sacrificial offering of weapons in the Thorsberger Moor took place at the transition from the Older to the Younger Roman Iron Age (B2/C1). This equipment originated from warriors from continental regions. A last deposit with artefacts from an eastern Scandinavian provenance took place in AD 300. For new studies regarding chronology and chorological questions in the Thorsberger Moor, the collections of so-called personal equipment of the warriors became the focus of the investigations (Blankenfeldt in prep.). These include dress 31 accessories, especially brooches and belt sets. The shape and design of these items were particularly susceptible to fashionable influences and the technical skills of the respective times. The representation of different tribes and tribal organizations through the distribution of dress accessories must always be critically questioned. Factors such as trade and migration patterns are certain to have an effect on the current distribution maps of specific objects and influence, largely unnoticed, the overall interpretation. Furthermore the informative value in the Thorsberger Moor remains severely limited due to undocumented correlations of findings. Unlike modern documentation of excavation, reliable information concerning objects found together is largely missing. In addition, various forms of jewellery, utilitarian objects, valuables and tools were combined under personal equipment. The research regarding the Militaria (Matešić in prep.) and horse harnesses (Lau in prep.) have a different focus of research interest, but also complement the chronological and chorological studies. Sacrifices in the Older Roman Iron Age in the Thorsberger Moor There are no indications of deposited military equipment from the Older Roman Iron Age in the Thorsberger Moor. A comparison with other large offering sites also shows that deposits of weapons from the Older Roman Iron Age are almost completely absent. Exceptions are Ejsbøl in Jutland and Vimose on Funen (Ilkjær 1990a: 272, 332, fig. 201; 2003a: 46, fig. 2; Andersen 2003; Pauli Jensen 2003: 227 f.). Seven Rollenkappenfibeln of different types represent artefacts from this period in the Thor sberger Moor (Raddatz 1957: Tab , 4 9; 1987b: cat. no. G261; ). These are interpreted by the author to be sacrifices of indigenous dress accessories (Blankenfeldt 2013: 63 f.; in prep.). According to another brooch from the late second early first century BC (Raddatz 1957: table 15. 3; 1987b: cat. no. G262), this fibula sacrifices probably spans an even larger chronological time frame. Comparisons with the burial and fibula sequence of the nearby graveyard Süderbrarup-am Markt (Bantelmann 1988) indicate similarities, which also infer relationships between the popula- 5 More detailed project descriptions: (Abegg-Wigg 2005; Abegg-Wigg et al. 2006; Carnap-Bornheim, Ilkjær 2008). The find material represented the basis for three PhD projects and further studies, these publications are currently in editorial revision: (Blankenfeldt in prep.; von Carnap-Bornheim in prep.; Lau in prep.; Matešić in prep).

32 32 Ruth BLANKENFELDT tions that were sacrificing here and burying there (see Rau 2010: 79 f.). Offerings of brooches are generally regarded as being sacrifices of female worshippers (Jud 2006: 97; see also Torbrügge 1972: 99, 119; Teegen 1999). Although offerings of fibulae by men generally can not be excluded, none have been documented as of yet (Jud 2006: 99). The utilization of the area now known as the Thorsberger Moor as a place to offer sacrifices as early as the Pre-Roman Iron Age is certain (Raddatz 1970; Willroth 1992: 223 f., 417 f.; Müller-Wille 1999: 49 f.). Evidence of this is a large variety of pottery and wooden vessels, as well as agricultural tools, which lead to the conclusion of a votive cult characterized by agriculture (Engelhardt 1893: Tab. 16, 17; Raddatz 1987a: cat. no , Tab ; Neben 2004). Thus, the deposits of vessels and brooches span a similar time period, even though the course of intensity can be described as diametrical based upon the findings available today. The objects deposited in the Thorsberger Moor in the Older Roman Iron Age are exclusively native dress accessories elements and ceramics. The character and probable intent of these sacrifices is significantly different from the symbolism connected to the deposit of spoils of war. Sacrifices during the Younger Roman Iron Age in the Thorsberger Moor A total of three deposits of military equipment in the Thorsberger Moor have been performed during the Younger Roman Iron Age (fig. 4). 6 A large percentage of the findings seem to belong to two sacrifices, both of which were performed in the C1 period. A definite association to one of these two deposits can t be derived for all of these objects, as they often consist of long-lasting forms. The definition of an earlier and later sacrifice in the C1 period can further be supported with the help of militaria (see Ilkjær 1993a: 283; Iversen 2008b: appendix II. 22; Matešić in prep.). In contrast, the horse harnesses from this time frame can be attributed to an advanced stage of period C1 (see Lau 2007; 2008; in prep.). The first offering of military equipment took place in the beginning Young Roman Iron Age (fig. 4, left). The find provenances indicate a large geographical area in continental Europe (fig. 4, left, light green). In addition to strong evidence of the Elbian-Germanic cultural area, there are many artefacts that are probably of an East-Germanic origin (fig. 4, left, dark green). The interpretation of such findings alongside the mixing of different cultural groups is questionable. Thus, alliances between men of different cultural groups are just as possible as the representation of a geographical area that lay in an overlapping zone of both influences. Probably the most comprehensive deposit of spoils of war happened in the first half of the third century AD (fig. 4, centre). A majority of the items found in the Thorsberger Moor could be attributed to this deposit (Ilkjær 1993a: 283; Biborski, Ilkjær 2006a: 383; Lau 2007; 2008; in prep.; Iversen 2008b: appendix II.22; Blankenfeldt in prep.; Matešić in prep.). An offering previously described in the literature as being the main deposit of period C1b is therefore verified; however, the earlier described extreme quantitative preponderance of 95 % of the overall findings from the Thorsberger Moor must be relativized (see Ilkjær, Lønstrup 1982: 95; Lønstrup 1984: 92). Elbian-Germanic find provenances are strongly represented in this offering, however North German and Danish, as well as Jutland components can also be noted (fig. 4, centre, light red). Such an observation is partially explained by alliances between units of differing origin, on the other hand, cultural parallels are also possible. This is for example observed on similarities in the Funen and Holstein form-spectrum. Due to significant individual artefacts and their typical pattern of distribution, the warriors potential regions of origin can be narrowed down a bit more clearly (fig. 4, centre, dark red). The final deposit of army equipment in the Thorsberger Moor took place at the early fourth century AD (fig. 4, right). The findings are comparable to deposits in other sacrificial sites, and therefore can be easily narrowed down chronologically (Ilkjær 1990a: 303 f., 334, fig. 203). Besides personal equipment artefacts, the functional groups horse harnesses (Lau 2008: 34 f.) and militaria (Ilkjær 1990a: 303 f., 334, fig. 203; Biborski, Ilkjær 2006a: 383; Iversen 2008b: appendix II. 22; Matešić in prep.) are also represented in this final deposit. Scandinavian components dominate in these findings (fig. 4, right, light blue) and suggest an origin of the warriors from the south and south-eastern area of Sweden and the Danish island regions (fig. 4, right, dark blue). 6 To maintain clearness, a detailed list of the objects, which were considered to be a chronological indicator, is not included here. For more information, see (Blankenfeldt in prep.).

33 Weapon Deposits in the Thorsberger Moor 150 Years of Research and New Рerspectives 33 Fig. 4. Chronology of Scandinavian lance- and spear-heads from AD (Ilkjær 2003b: 29, fig. 21)

34 34 Ruth BLANKENFELDT Fig. 5. Destruction occurred in a systematic manner: ritual treatment of snake head bracelets from the Thorsberger Moor (photos: Regional Museum of Archaeology of Schleswig) The thorsberger Moor: militaria and horse harnesses Weapons are a central functional group represented in deposits of military equipment. Due to the acid ph-value in the Thorsberger Moor, the quota of loss in this object group can certainly be classified as very high. The transmitted metal artefacts consisting of copper alloys and precious metals are mostly attributable to sword and shield sets. Here a relatively high proportion of objects is recognizably of Roman origin. If a statement can be made regarding the provenance of these finds, these mainly refer to the provinces of Germania Inferior and Britannia (Matešić in prep.). Among the militaria there are some striking individual pieces, for example a silver face mask (Engelhardt 1863: table 5. 3; Raddatz 1987a: cat. no. 405, tab. 90.1; 91.1; fig. 23). During a revision of the finds from the Thorsberger Moor, S. Matešić could explain that this piece, along with other metal artefacts found here, could be reconstructed into a complete helm, originally with an iron and therefore meanwhile gone helmet calotte (Matešić 2010; in prep.). Models for this Germanic-made and so far unique helmet are to be found in Roman pieces. Parallels to mask helmets of the female type from the second half of the second and early third century AD, which are primarily known from the Danube region, become particularly clear (Matešić 2010: 56 f., fig. on p. 57; in prep.). In the Thorsberger Moor, an unusually high number of Germanic Roman Iron Age horse harness pieces could be, which N. Lau could assemble into at least two dozen sets (Lau 2007; 2008; in prep.). The majority of them were probably sacrificed along with the main deposit in the first half of the third century AD. At least nine sets ended up in the Thorsberger Moor in the final sacrifice (see above). Besides models from the Roman empire combined with Germanic design and manufacturing styles, the Germanic horse harnesses prove contact to further cultural areas. Thus, there are indications of parallels with Sarmatian and Pannonian horse harness pieces, as well as with the North Danubian area at the time of the Marcomannic Wars (Lau in prep.). Ritual destruction The Germanic harnesses and bridle chains are often richly decorated and superior in quality. They can be associated with a higher status of the riders. This lead to the conclusion that members of the military command in particular, but also the social elite, were mounted and participated directly in battle events (Lau in prep.). On the other hand,

35 Weapon Deposits in the Thorsberger Moor 150 Years of Research and New Рerspectives these mounted warriors are just a few and the riding units in the Germanic world are not comparable with big cavalry units like, for example, in the Roman empire. In addition, the objects in this function group show a high proportion of ritually induced destruction (Lau 2009). The representation of a high status by using insignia made of precious raw material played an important role in the Germanic world. The special treatment of status symbols is often observed in the weapon deposits. Thus, a high degree of destruction can be observed on many insignia in addition to that on the horse harnesses. This is obvious, for example, on the basis of the snake head bracelets from the Thorsberger Moor, which had been unrolled and cut from the back prior deposition (Raddatz 1957: Tab. 19; Andersson 1993: 126 f., cat. no a j; Blankenfeldt 2013: 59) (fig. 5). In addition, deep notches and grooves are recognizable on the front side of the ring fragments, which also attest to destruction that had been intensively carried out. The ritual destruction, especially of objects that indicate status, is certainly also to be interpreted as a symbolic act of power devaluation. Clear analogies are recognizable in relation to the treatment and deposit military equipment between the Northern European sacrificial sites. The destruction of the items was carried out very carefully, 35 in a predetermined manner and followed a specific pattern in the selection of items and the kind of the destruction. These observations argue against uncontrolled, frenzied sacrificial ceremonies, such as were described by Orosius (see above). The deposit of the spoils of war are rather to be understood as a controlled presentation of triumph, which may have been adapted from models of antiquity like from the Celtic and Roman cultural areas (Jørgensen 2001: 14 f., fig. 7 8; Blankenfeldt, Rau 2009: 134). A demonstration of sovereign legitimation in this form also represents a religious and political power element that was unanimous in the Northern European realm. It was exercised similarly in spite of the numerous Germanic internal conflicts. *** The analysis of the sacrifice sites with weapon deposits touches on very different subject areas and archaeological issues. In addition to chronological and chorological studies, the extensive find ensembles also allow overriding interpretations. These include statements regarding tactics, size and organization of the Germanic army units, as well as transfers of ideas and technology, as well as transregional cultural connections. After all, the sacrificial practices tie together the areas of religious worship and the demonstration of political power for the Germanic people. Bibliography Abegg-Wigg A. Zwischen Thorsberg und Bornstein. Zu einem neuen Forschungsprojekt am Archäologischen Landesmuseum in Schleswig // Archäologisches Nachrichtenblatt Abegg-Wigg A., Blankenfeldt R., Lau N., Matešić S. Zwischen Thorsberg und Bornstein // Arkæologi i Slesvig Aktuelle Forschungen zu Kriegsbeuteopfern und Fürstengräbern im Barbaricum. Internationales Kolloquium unterstützt durch Carlsbergfondet, Schleswig Juni 2006 / Schriften des Archäologischen Landesmuseums. Ergänzungsreihe. 4. Neumünster, Altes Germanien. Auszüge aus den antiken Quellen über die Germanen und ihre Beziehungen zum Römischen Reich. Quellen der Alten Geschichte bis zum Jahre 238 n. Chr. 1. Darmstadt, Andersen H. C. Neue Untersuchungen im Moor von Ejsbøl // Sieg und Triumpf. Der Norden im Schatten des Römischen Reiches. Kopenhagen, Bantelmann N. Süderbrarup. Ein Gräberfeld der römischen Kaiserzeit und Völkerwanderungszeit in Angeln 1. Archäologische Untersuchungen / Offa-Bücher. 63. Neumünster, Bemmann J., Hahne G. Waffenführende Grabinventare der jüngeren römischen Kaiserzeit und Völkerwanderungszeit Skandinaviens. Studie zur zeitlichen Ordnung anhand der norwegischen Funde // BRGK Bemmann J., Bemmann G. Der Opferplatz von Nydam. Die Funde aus den älteren Grabungen: Nydam-I und Nydam II 1. Text. Neumünster, 1998a. Bemmann J., Bemmann G. Der Opferplatz von Nydam. Die Funde aus den älteren Grabungen: Nydam-I und Nydam II 2. Katalog und Tafeln. Neumünster, 1998b.

36 36 Ruth BLANKENFELDT Biborski M., Ilkjær J. Illerup Ådal. 11. Die Schwerter. Textband / Jutland Archaeological Society Publications. 25/11. Aarhus, 2006a. Biborski M., Ilkjær J. Illerup Ådal. 12. Die Schwerter. Tafelband / Jutland Archaeological Society Publications. 25/12). Aarhus, 2006b. Blankenfeldt R. Das gebogene Blech aus dem Thorsberger Moor // Aktuelle Forschungen zu Kriegsbeuteopfern und Fürstengräbern im Barbaricum. Internationales Kolloquium unterstützt durch Carlsbergfondet, Schleswig Juni / Schriften des Archäologischen Landesmuseums. Ergänzungsreihe. 4. Neumünster, Blankenfeldt R. Überlegungen zu Hinweisen auf Individualität in Deponierungen von Heeresausrüstungen // Individual and Individuality. Approaches towards an Archaeology of Personhood in the First Millennium AD / Neue Studien zur Sachsenforschung. 4. Hannover, Blankenfeldt R. Die persönlichen Ausrüstungen // Thorsberger Moor. 1 (in prep.). Blankenfeldt R., Rau A. Skandinavische Kriegsbeuteopfer Befunde, Funde und Interpretation. In: 2000 Jahre Varusschlacht: Konflikt. Stuttgart, Bursche A. Illerup Ådal. 14. Die Münzen / Jutland Archaeological Society Publications. 25/14. Aarhus, Carnap-Bornheim C. von. Neue Forschungen zu den beiden Zierscheiben aus dem Thorsberger Moorfund // Germania /1. Carnap-Bornheim C. von. Forschungsgeschichte, naturwissenschaftliche Untersuchungen und Konkordanzlisten // Thorsberger Moor. 4 (in prep.). Carnap-Bornheim C. von, Ilkjær J. Illerup Ådal. 5. Die Prachtausrüstungen. Textband / Jutland Archaeological Society Publications. 25/5. Aarhus, 1996a. Carnap-Bornheim C. von, Ilkjær J. Illerup Ådal. 6. Die Prachtausrüstungen. Katalog, Fundlisten und Literatur / Jutland Archaeological Society Publications. 25/6. Aarhus, 1996b. Carnap-Bornheim C. von, Ilkjær J. Illerup Ådal. 7. Die Prachtausrüstungen. Tafelband / Jutland Archaeological Society Publications. 25/7. Aarhus, 1996c. Carnap-Bornheim C. von, Ilkjær J. Illerup Ådal. 8. Die Prachtausrüstungen. Grabungsdokumentation und Fundliste / Jutland Archaeological Society Publications. 25/8. Aarhus, 1996d. Carnap-Bornheim C. von, Ilkjær J. Jernalderen i Nordeuropa und Zwischen Thorsberg und Bornstein Archäologische Forschung im internationalen Netzwerk // Aktuelle Forschungen zu Kriegsbeuteopfern und Fürstengräbern im Barbaricum. Internationales Kolloquium unterstützt durch Carlsbergfondet, Schleswig Juni 2006 / Schriften des Archäologischen Landesmuseums. Ergänzungsreihe. 4. Neumünster, Carnap-Bornheim C. von, Rau A. Zwischen religiöser Zeremonie und politischer Demonstration Überlegungen zu den südskandinavischen Kriegsbeuteopfern // Glaube, Kult und Herrschaft. Phänomene des Religiösen im 1. Jahrtausend n. Chr. in Mittel- und Nordeuropa / Kolloquium zur Vor- und Frühgeschichte. 12. Bonn, Eggers H.-J. Der römische Import im Freien Germanien. Atlas der Urgeschichte. 1. Hamburg, Eggers H.-J. Zur absoluten Chronologie der Römischen Kaiserzeit im Freien Germanien // Jahrbuch des RGZM Engelhardt C. Thorsbjerg Mosefund. Sønderjydske Mosefund. 1. Kjöbenhavn, Engelhardt C. Nydam Mosefund. Sønderjydske Mosefund. 2. Kjøbenhavn, Engelhardt C. Denmark in the Early Iron Age illustrated by recent discoveries in the peat mosses of Slesvig. London, Engelhardt C. Kragehul Mosefund. Fynske Mosefund. 1. København, Engelhardt C. Vimose fundet. Fynske Mosefund. 2. Kjøbenhavn, Ilkjær J. Et bundt våben fra Vimose // Kuml Ilkjær J. Illerup Ådal. 1. Die Lanzen und Speere. Textband / Jutland Archaeological Society Publications. 25/1. Aarhus, 1990a. Ilkjær J. Illerup Ådal. 2. Die Lanzen und Speere. Tafelband / Jutland Archaeological Society Publications. 25/2. Aarhus, 1990b. Ilkjær J. Illerup Ådal. 3. Die Gürtel Bestandteile und Zubehör. Textband / Jutland Archaeological Society Publications. 25/3. Aarhus, 1993a.

37 Weapon Deposits in the Thorsberger Moor 150 Years of Research and New Рerspectives 37 Ilkjær J. Illerup Ådal. 4. Die Gürtel Bestandteile und Zubehör. Tafelband. / Jutland Archaeological Society Publications. 25/4. Aarhus, 1993b. Ilkjær J. Illerup Ådal. 9. Die Schilde. Textband / Jutland Archaeological Society Publications. 25/9. Aarhus, 2001a. Ilkjær J. Illerup Ådal. 10. Die Schilde. Tafeln und Fundlisten / Jutland Archaeological Society Publications. 25/10. Aarhus, 2001b. Ilkjær J. Illerup Ådal Ein archäologischer Zauberspiegel. Højbjerg, Ilkjær J. Dänische Kriegsbeuteopfer // Sieg und Triumpf. Der Norden im Schatten des Römischen Reiches. Kopenhagen, 2003a. Ilkjær J. Mosens Skatkammer. Mellem mennesker og guder i jernalderen. Mosgård, 2003b. Ilkjær J., Lønstrup J. Interpretation of the Great Votive Deposits of Iron Age weapons // Journal of Danish Archaeology Ilkjær J., Iversen R. Untergegangen. Germanische Heeresverbände und skandinavische Kriegsbeuteopfer. In: 2000 Jahre Varusschlacht: Konflikt. Stuttgart, Iversen R. Fourth and Fifth Century War Booty from Kragehul // Aktuelle Forschungen zu Kriegsbeuteopfern und Fürstengräbern im Barbaricum. Internationales Kolloquium unterstützt durch Carlsbergfondet, Schleswig Juni 2006 / Schriften des Archäologischen Landesmuseums Ergänzungsreihe. 4. Neumünster, 2008a. Iversen R. Kragehul et krigsbytteofferfund på Sydvestfyn. Kronoligiske og typologiske bidrag til forstålsen af krig og samfund i yngre romersk jernalder og folkevandringstid i Skandinavien (Dissertation). Århus, 2008b. Iversen R. Kragehul mose. Ein Kriegsbeuteopfer auf Südwestfünen / Jysk Arkæologisk Selskab. 73. Århus, Jankuhn H. Zur Deutung des Moorfundes von Thorsberg // Forschungen und Fortschritte. 1936a. 12. Jankuhn H. Die religionsgeschichtliche Bedeutung des Thorsberger Fundes // Forschungen und Fortschritte. 1936b. 12. Jud P. Küche, Kinder Kult? Die Rolle der Frau in den Kulten der alpinen und nordalpinen Eisenzeit: Versuch einer kritischen Bestandaufnahme // Studien zur Lebenswelt der Eisenzeit / Reallexikon Der Germanischen Altertumskunde. Erganzungsband 53. Berlin, New York, Jørgensen L. The Warriors, Soldiers and Conscripts of the Anthropology in Late Roman and Migration Period Archaeology // Military Aspects of the Aristocracy in Barbaricum in the Roman and Early Migration Periods. Copenhagen, Lau N. Die Pferdegeschirre aus dem Thorsberger Moor Neue Forschungen zu den Ausrüstungen der germanischen Reiterei // Innere Strukturen von Siedlungen und Gräberfeldern als Spiegel gesellschaftlicher Wirklichkeit? Akten des 57. Internationalen Sachsensymposions vom 26. bis in Münster. Münster, Lau N. Zügelkettenzaumzeuge der jüngeren und späten Römischen Kaiserzeit Neue Untersuchungen zu Typen, Verbreitung, Herkunft und Datierung // Aktuelle Forschungen zu Kriegsbeuteopfern und Fürstengräbern im Barbaricum. Internationales Kolloquium unterstützt durch Carlsbergfondet, Schleswig Juni 2006 / Schriften des Archäologischen Landesmuseums Ergänzungsreihe. 4. Neumünster, Lau N. Zum Ritus der Opferung von Kriegsbeute in der jüngeren römischen Kaiserzeit Spuren ritueller Zerstörungen an Pferdegeschirren aus dem Thorsberger Moorfund. // Waffen in Aktion. Akten der 16. Internationalen Roman Military Equipment Conference (ROMEC), Xanten, Juni 2007 / Xantener Berichte. 16. Mainz am Rhein, Lau N. Die Pferdegeschirre. Studien zu germanischen Zaumzeugen und Sattelgeschirren als Zeugnisse der militärischen Reiterei im mittel- und nordeuropäischen Barbaricum // Thorsberger Moor. 2 (in prep.). Lønstrup J. Older and newer Theories. The find from Thorsbjerg in the light of new discoveries // Frühmittelalterlichen Studien Lund Hansen U. Römischer Import im Norden. Warenaustausch zwischen dem Römischen Reich und dem freien Germanien // Nordiske Fortidsminder. Ser. B. 10. København, Lund Hansen U. 150 Jahre Waffenopferfunde Forschung und Interpretation // Sieg und Triumpf. Der Norden im Schatten des Römischen Reiches. Kopenhagen, Matešić S. Der germanische Helm im Thorsbeger Moor // Archäologische Nachrichten aus Schleswig-Holstein Matešić S. Die militärischen Ausrüstungen // Thorsberger Moor. 3 (in prep.). Müller-Wille M. Opferkulte der Germanen und Slawen. Stuttgart, 1999.

38 38 Ruth BLANKENFELDT Neben M. Die Ton- und Holzgefäße aus dem Thorsberger Moor, Gem. Süderbrarup, Kr. Schleswig-Flensburg (Magisterarbeit). Kiel, Ørsnes M. Der Moorfund von Ejsbøl bei Hadersleben und die Deutungsprobleme der großen germanischen Waffenopferfunde // Vorgeschichtliche Heiligtümer und Opferplätze in Mitteleuropa. Kongreß Rheinhausen 1968 / Abhandlungen der Akademie der Wissenschaften zu Göttingen, Historisch-Philologische Klasse. 3/74. Göttingen, Ørsnes M. Sejrens pris. Våbenofre i Ejsbøl Mose ved Haderslev. Haderslev, Ørsnes M. Ejsbøl I, Waffenopferfunde des Jahrhunderts nach Chr. / Nordiske Fortidsminder. Ser. B. 11. København, Pauli Jensen X. Der Moorfund aus Vimose // Sieg und Triumpf. Der Norden im Schatten des Römischen Reiches. Kopenhagen, Pauli Jensen X., Nørbach L. C. Illerup Ådal. 13. Die Bögen, Pfeile und Äxte / Jutland Archaeological Society Publications. 25/13. Aarhus, Raddatz K. Bronzezeitliche Funde aus dem Thorsberger Moor // Jahrbuch der Heimatvereins der Landschaft Angeln Raddatz K. Der Thorsberger Moorfund. Gürtelteile und Körperschmuck / Offa-Bücher. 13. Neumünster, Raddatz K. Zu den magischen Schwertanhängern des Thorsberger Moorfundes // Offa. 1957/ Raddatz K. Religionsgeschichtliche Probleme des Thorsberger Moorfundes // Vorgeschichtliche Heiligtümer und Opferplätze in Mitteleuropa. Kongreß Rheinhausen 1968 / Abhandlungen der Akademie der Wissenschaften zu Göttingen, Historisch-Philologische Klasse. 3 / 74. Göttingen, Raddatz K. Der Thorsberger Moorfund. Katalog. Teile von Waffen und Pferdegeschirr, sonstige Fundstücke aus Metall und Glas, Ton- und Holzgefäße, Steingeräte / Offa-Bücher. 65. Neumünster, 1987a. Raddatz K. Der Thorsberger Moorfund. Gürtelteile und Körperschmuck // Offa. 1987b. 44. Rau A. Nydam mose 1 2. Die personengebundenen Gegenstände. Grabungen Jernalderen i Nord europa / Jysk Arkæologisk Selskab. 72. Aarhus, Rau A. Weihungen von Kriegerausrüstungen bei den nördlichen Germanen der Kaiser- und Völkerwanderungszeit // Waffen für die Götter. Krieger Trophäen Heiligtümer. Innsbruck, Rau A., Carnap-Bornheim C. von. Die kaiserzeitlichen Heeresausrüstungsopfer Südskandinaviens Überlegungen zu Schlüsselfunden archäologisch-historischer Interpretationsmuster in der kaiserzeitlichen Archäologie // Altertumskunde Altertumswissenschaft Kulturwissenschaft. Erträge und Perspektiven nach 40 Jahren Reallexikon der Germanischen Altertumskunde / Reallexikon Der Germanischen Altertumskunde. Erganzungsband 77 (2011). Berlin; Boston, Sieg und Triumpf. Der Norden im Schatten des Römischen Reiches. Kopenhagen, Teegen W.-R. Studien zu dem kaiserzeitlichen Quellopferfund von Bad Pyrmont // Reallexikon Der Germanischen Altertumskunde. Ergänzungsbande 20. Berlin, New York, Torbrügge W. Vor- und Frühgeschichtliche Flussfunde. Zur Ordnung und Bestimmung einer Denkmälergruppe // BRGK /52. Werner J. Die beiden Zierscheiben des Thorsberger Moorfundes. Ein Beitrag zur frühgermanischen Kunst- und Religionsgeschichte. Berlin, Wiell S. Der dänische Moorfundpionier Der Archäologe Conrad Engelhardt und sein Schaffen // Sieg und Triumpf. Der Norden im Schatten des Römischen Reiches. Kopenhagen, Willroth K.-H. Untersuchungen zur Besiedlungsgeschichte der Landschaften Angeln und Schwansen von der älteren Bronzezeit bis zum frühen Mittelalter: eine Studie zur Chronologie, Chorologie und Siedlungskunde / Offa Bücher. 72. Neumünster, Worsaæ J. J. A. Om Slesvigs eller Sønderjyllands oltidsminder. Kjøbenhavn, 1865.

39 Weapon Deposits in the Thorsberger Moor 150 Years of Research and New Рerspectives 39 Рут БЛАНКЕНФЕЛЬДТ Комплексы с оружием в «болоте Торсберг» 150 лет изучения и новые перспективы Резюме Болото Торсберг находится в Зюдербрарупе на севере Германии, земля Шлезвиг-Гольштейн, недалеко от датской границы. В гг. Х. К. Энгельхардт расчистил 750 м 2 отно сительно сухого болота. Он обнаружил множество археологических находок и отметил ряд их характерных особенностей. В частности, примечательно, с одной стороны, большое количество предметов, открытых на небольшой площади (рис. 2), а с другой стороны то, что большинство находок, например, щитов, мечей, пик и копий указывают на существенный военный компонент среди обнаруженных материалов. Он отметил также, что на вещах сохранились повреждения, которые в некоторых случаях появились в результате более ранних военных действий. Кро ме того, ряд обстоятельств указывает на систематический характер повреждений. В целом, по заключению Х.-К. Энгельхардта, вещи оказались на этих памятниках преднамеренно. Он был первым, предложившим называть такие памятники «местами приношений». Со времени раскопок Х.-К. Энгельхардта в болоте Торсберг и сопоставимых датских памятниках прошло более 150 лет. Сегодня известно более двух десятков подобных памятников, «мест, где приносили в жертву во енные трофеи (Kriegsbeuteopferplätze)», или «мест, где приносили в жертву военное снаряжение» (рис. 1). Их следует ин тер пре тировать как местности, где приносили в жертву немногочисленные, но чрезвычайно богатые комплексы военных трофеев, взятых германскими племенами. В этих комплексах было оружие и, например, большое количество деталей конской упряжи, а также разнообразные вещи, сделанные из органических материалов. Одновременные комплексы, состоящие из сотен подобных предметов, представляют собой явление, близкое приношениям в болотах. Большое количество синхронного оружия, най денного на памятниках с приношениями, в сочетании с погребальными памятникам этого времени дают основания описать от дель ные типы, обладающие фиксированными мор фологическими характеристиками, образуя ос нову для хронологии Северной Европы (рис. 4). Сами вещи интерпретируют как объекты не местного происхождения и не принадлежавшие тем, кто совершал жертвоприношения. Веро ят но, их отбирали у воинов из других гео графических областей. Помимо черт, отсы лающих к битвам прошлого, на некоторых на ходках сохранились черты ритуальных действий, пред принятых до того, как их принесли в жертву (рис. 5). В целом, эти места дают ценную информацию о живой культуре север ного Барбарикума в большей степени, чем дру гие археологические находки. Речь идёт о сра жениях, жертвоприношениях и связанных с ни ми церемония, а кроме того, о демонстрации по ли тической власти. Современные исследования шлезвигского Цент ра Балтийской и скандинавской археологии Шле з вига (Готторфский замок) направлены на изучение более чем 2500 находок из болота Торсберг современными методами. В конце доримского железного века и в начале римского времени болото Торсберг было местом, где совершали приношения керамической и деревянной посуды, а также фибул. В начале римского времени можно выделить три случая, когда в жертву приносилось воинское снаряжение (рис. 3). Первое приношение военных трофеев произошло на рубеже доримского и римского времени. Воины, совершавшие прино шения в это время, происходили из континен тальной Европы, в том числе из областей, за селённых эльбскими германцами, и с тер ри тории пшеворской культуры (рис. 3, слева). Ко личество военных трофеев, оказавшихся в бо лоте Торсберг, достигло своего максимума в гг. н. э. Несмотря на доминирование нахо док, связанных с эльбскими германцами, в комп лексах этого хронологического периода от ме чены компоненты, происходившие из север ной Германии, Ютландии и датских ост ровов (рис. 3, в центре). Последнее при ношение в болоте Торсберг произошло в гг. В ге о- графическом отношении, эти воины, вероятно, происходили из южной и юго-восточной Швеции и из района датских островов (рис. 3, справа).

40 40 Linda Boye, Ulla Lund Hansen Glass and Amber Beads in Late Roman Iron Age. Relations between Denmark and the Black Sea Area with a special focus on the graves from Eastern Zealand Introduction to the Eastern Zealand in Late Roman Iron Age During the Late Roman Iron Age and Migration period contacts between Scandinavia, especially the southern part of it, and Southeast Europe has been clearly demonstrated by mean of groups of objects especially the many drinking glass vessels. But to this far-reaching contact must also be added beads. In Scandinavian and other North European grave finds beads are seldom in graves from the Early Roman Iron Age B1 B2, and when beads now and then occur it is always in rich and wellequipped female graves. From the beginning of the Late Roman Iron Age beads became frequent in female graves and at central sites as well (Lund Hansen 2009b: 90 99; 2009c). During most of the Late Roman Iron Age the stock of beads are considerable and have great variety. Very often the richness of the female graves is combined with exclusivity of the beads. Around the beginning of phase C3 the glass bead material maybe the bead fashion maybe the access to glass beads changed in both South Scandinavia and in Barbaric Europe in general. In South Scandinavia the number of beads in average per grave increased, but at the same time the type of glass beads changed and the variety of glass bead types now became limited. Den Rige Vestegn is a research area west of Copenhagen, where during the past 30 years nine burial sites, dated to the Late Roman Iron Age periods C1b to C3, have been excavated (fig. 1). Together with the graves traces of hundreds of structures belonging to different kinds of settlements were discovered in the area. In recent years one has gained an insight into the relation between related settlements and burial sites and also into contemporary burial sites. This has created a unique opportunity to study Late Roman Iron Age society from supplementary sources which might give a more real image of society development. During the Late Roman Iron Age there existed three contemporary house types on eastern Zealand: the Høje-Taastrup type, the Ragnesminde type, and the Stuvehøj type. Each of these types can be linked to a distinctive class or level in society. The house types did either occurred together in the same settlements, or the Ragnesminde farms and the Stuvehøj farms were situated as single farm units surrounded by their fields (Boye 2008) (fig. 2 3). In the region of the Den Rige Vestegn isolated farms and dense villages occurred side by side. The local lord or princeps and the upper class of the society lived in characteristic houses, and it is likely that the princeps lived in the largest farm with fields surrounding the structure during the Late Roman Iron Age. This very house type is of the so-called Stuvehøj type and can, for example at Brøndsager, be observed approximately 200 m from the rich graves (fig. 4). Warriors in service of the princeps lived at the nearby Ragnesminde farms. Many examples have proved that a Ragnesminde structure is situated approximately 200 meter from rich graves, this is the case at Torstorp Vesterby, Stenrøldsknøs, Engbjerg and Ryget Skovby in Værløse, and at Højbakkegård (Boye 2008). General aspects of the inhumation graves Burial mounds are in this region not documented by remains of actual mounds. However, by examining the location of graves on the burial sites, it can be shown that certain graves most likely originally were covered by a mound (examples are Torstorp

41 Glass and Amber Beads in Late Roman Iron Age. Relations between Denmark and the Black Sea Area with a Special Focus on the Graves from Eastern Zealand 41 Fig. 1. The research area West of Copenhagen, where now nine burial sites from Late Roman Iron Age are found (map: Freerk Oldenburger, Kroppedal Museum, remade by Maxim Levada)

42 42 Linda Boye, Ulla Lund Hansen Fig. 2. The Ragnesminde Farms. This house type can be linked to people from the burial sites, who were buried with Roman imports (drawing: Kroppedal Museum)

43 Glass and Amber Beads in Late Roman Iron Age. Relations between Denmark and the Black Sea Area with a Special Focus on the Graves from Eastern Zealand 43 Fig. 3. The Stuvehøj Farms. This house type is greater in both length and width than the Ragnesminde Farm and it can be linked to the local prince (drawing: Kroppedal Museum)

44 44 Linda Boye, Ulla Lund Hansen Fig. 4. At the site Brøndsager the Stuvehøj house type can be observed approximately 200 m from the rich princeps grave (drawing: Charlotte Clante, Kroppedal Museum)

45 Glass and Amber Beads in Late Roman Iron Age. Relations between Denmark and the Black Sea Area with a Special Focus on the Graves from Eastern Zealand 45 Vesterby, Brøndsager, Stenrøldsknøs, Engbjerg, and Ellekilde). The burials are orientated in north-south direction. In ordinary graves the deceased were laid with the head to the north. Furthermore the position of the body was determined by gender males were laid on their back and females on their side with the face towards the east (fig. 5). In the richest graves, such as Brøndsager and Ellekilde, the deceased was buried with the head to the south. Large stones were often placed on top of the richest graves, especially those in which the deceased was male. The graves of middle class men were generally poor in grave goods as most of their possessions maybe were inherited by descendants (fig. 6). The Late Roman Iron Age burials contained a variety of grave goods linked to the social status of the deceased and also food for the journey to the afterlife. A regular burial custom was the symbolic payment to the ferryman Charon. At the Engbjerg burial site it was recognized for the first time, that that kind of payment not necessarily had to be a coin or a piece of gold but also could be shards from Roman drinking glasses (Boye 2002) (fig. 7). Subsequently it has been realized, that the custom of laying pieces of glass in the grave not only was limited to the area West of Copenhagen but also could be found elsewhere at Zealand. The glass shards were a symbolical representation of Roman drinking glasses with which the ferryman was to be paid. The nine burial sites Torstorp Vesterby (fig. 8). In 1989 a small burial site dated to the Late Roman Iron Age C3 was uncovered and excavated by curator Eliza Fonnesbech-Sandberg; it consisted of eight inhumation graves one wealthy adult woman and seven children. The seven small graves were aligned almost exactly north-south, while the richly endowed adult grave laid NNW SSW. The children s graves were all situated north of the adult grave, and they were Fig 5. Illustration showing the typical way of bury women and men: Engbjerg grave 18, woman, in sleeping position on her side, and Engbjerg grave 8, man, lying on his back (drawing: Kroppedal Museum)

46 46 Linda Boye, Ulla Lund Hansen Fig. 6. Engbjerg grave 8, male grave covered with big stones (drawing: Kroppedal Museum)

47 Glass and Amber Beads in Late Roman Iron Age. Relations between Denmark and the Black Sea Area with a Special Focus on the Graves from Eastern Zealand 47 Fig. 7. Engbjerg grave 4, female grave with a glass shard from a Roman glass bowl laid by the woman s mandible (photo: Kroppedal Museum) either empty (two) or contained a single coarse ware jar (two) or a coarse ware jar and an amber bead. Two half-length graves included three respectively ten objects. Remarkable was grave 3368 which contained the remains of a woman aged approximately 25. Brøndsager (fig. 9) consists of three inhumation graves: an adult man, a young man and a little girl. The site was excavated headed by curator Eliza Fonnesbech-Sandberg and curator Linda Boye. The graves were grouped close together and may have been covered by one or several burial mounds. The most important grave contained the skeleton of a young man/boy of approximately 12 years. He was lying in an extended supine position with his head towards the south. Around his neck he wore a necklace of glass and amber beads, several bucket-shaped pendants and a gold aureus with a loop. In his mouth was placed a cutting of gold thread, which served as Charon s coin. On one of the fingers on his left hand lay a badge of rank: a gold snake s-head ring. Beside the head stood a wooden bucket held together with a bronze ring as well as two coarseware jars. Next to the jars were two fragments of glass vessels. In the middle of the grave was a gaming board with red and white fields along with black and white/blue-green glass gaming pieces. A bone comb was found beside the legs. Two more coarse-ware jars were found by the feet. In addition, the grave contained several animal offerings: a young sheep, parts of an adult cow and two pigs (one of which was a whole pig aged 3 9 months, the other months) as well as individual horse bones from a mature specimen probably symbolising a horse for the deceased to ride in the afterlife. Overall, the grave can be dated to the Late Roman Iron Age C2 (Boye 2009). Stenrøldsknøs consists of a single rich man s grave located close to a Late Neolithic burial mound. The grave was excavated in 1999 headed by archaeologist Henrik Høyer. The grave was found immediately below the subsoil so close to the surface, that the excavator spade revealed the

48 48 Linda Boye, Ulla Lund Hansen Fig. 8. Torstorp Vesterby grave 3368 from phase C3. Three necklaces, two of them made of amber beads and one made of yellow and red glass beads (photo: Erling Pultera, Kroppedal Museum)

49 Glass and Amber Beads in Late Roman Iron Age. Relations between Denmark and the Black Sea Area with a Special Focus on the Graves from Eastern Zealand 49 Fig. 9. Necklace made of 90 amber beads and two bucket shaped bronze pendants. Brøndsager grave 2769 from phase C1b C2, grave for a child aged eighteen months (photo: Kroppedal Museum) bronze handle of the Hemmoor bucket. The gender was determined as male solely on the basis of the grave goods. Traces of the deceased (in the ground) tell that he was cm tall. The grave goods consisted of a Roman dinner set in the form of a bronze dish, a Hemmoor bucket, two coarse-ware jars and two Roman drinking glasses. The deceased also wore a large spiral gold ring on his finger. Dating: Late Roman Iron Age C2 (Boye 2009). Engbjerg (fig. 10) consists of 25 inhumation graves for children, adult women and men. The site was excavated in 1998 headed by curator Linda Boye. Its topographical location is a hilltop 39 metres above sea level considerably higher

50 50 Linda Boye, Ulla Lund Hansen a b Fig. 10. Selection of necklaces from the grave field Engbjerg: a grave 4 (phase C2, woman age 20 40); b grave 15 (phase C1b2, age 5 6 months, sex unknown); c grave 24 (phase C1b2, age 18 months, sex unknown); d grave 10 (early phase D, age 20 40, sex unknown) (photo: Kroppedal Museum)

51 Glass and Amber Beads in Late Roman Iron Age. Relations between Denmark and the Black Sea Area with a Special Focus on the Graves from Eastern Zealand 51 c d

52 52 Linda Boye, Ulla Lund Hansen than the surrounding landscape. All graves were oriented north-south with the corpses heads towards the north. Most graves can be dated to the Late Roman Iron Age C1b to C2. One grave was younger (C3/D1). Grave 4 contained a particularly wealthy woman aged and 1.65 metres tall. The deceased wore a hairnet consisting of 54 glass and amber beads. These beads are small compared to those from the necklaces and consist predominantly of oblong and small, round, blue glass and amber beads as well as small, round, red, yellow, green, black, turquoise, and white glass beads. A silver hairpin fixed the hairnet. The type of pin is dated to late C1b and C2. A tight bead necklace was worn around the neck. It consisted of 63 amber beads: flat amber beads and big cylindrical amber beads of varying sizes arranged in a symmetrical and sequential fashion (small beads first). A bigger necklace (approximately 68 beads) consisting of very different glass beads was either worn around the neck or fastened to the breast. On the deceased s chest was placed a huge swastika fibula (Boye 2009). Kraghave Ødetofter (fig. 11) consists of three inhumation graves: two women and one man. The burial site was excavated in 2000 headed by archaeologist Anne B. Hansen. Three inhumation graves emerged during the excavation of a major settlement that primarily dates to the Late Iron Age and the Middle Ages. The graves lie on a conspicuous hillock, oriented N S and E W. Two women graves and a male grave were found. The oldest of the graves was for a man laid in an extreme foetal position and dated to the Early Roman Iron Age, while the two women graves are much younger from the Late Roman Iron Age C3. The inhumation graves were probably marked either with a small burial mound or with stones (Boye 2009). King Svends Park (fig. 12) consists of nine inhumation graves for one adult and eight children dated to C1b/C2 C3. The site was excavated in 2000 headed by Linda Boye (Boye 2009). The burial site is located 45 metres above sea level and is among the area s highest ancient monuments. To be remarked is that many items in the graves show connections or parallels to the Wielbark-culture in the north of Poland (Boye 2009; Cieslinski 2009). Højbakkegård (fig. 13) consists of a founder s grave dated to the Stone Age and six inhumation burials from the Roman Iron Age. The burial site was excavated in 2001 headed by curator Tom Giersing. The burial site contained a total of six inhumation graves from the Late Roman Iron Age located around a much older burial mound from the Late Neolithic period. Ellekilde consists of 30 inhumation graves from Late Roman Iron Age C1b/C2 to Migration period D1. The site was excavated in 2007 headed by curator Rune Iversen. The founder s grave was a princely chamber grave, which hold 61 black and white glass gaming pieces, a 10 cm long silver fibula with settings of semiprecious stone and glass, bronze mountings and silver rivets from a large bugle, a heavy gold finger-ring, a bone comb and a badly preserved glass shard representing the Charon s fee. The grave goods include Germanic as well as Roman tableware. The Germanic tradition is represented by the pottery: a jug, a cup and a foot-beaker, while the Roman dining set consists of a Hemmoor bucket, ladle and strainer and two painted glass beakers (circus cups). Around the male grave were situated several female graves (Iversen 2009) (fig. 14). Snubbekorsgård (fig. 15) is normally known as a Viking age cemetery, but has a Late Roman Iron Age founder s grave (dated to Late Roman Iron Age C1 C2) located in a burial mound. The site was excavated in headed by curator Lotte Reedtz Sparrevohn. There was no trace left of the deceased, but it was possible to conclude, that the dead woman has been placed with the head to the north because of the bead string with amber berloque-shaped beads, rounded amber beads and a few glass beads. Focusing on the female graves it occurred, that almost all the graves contained glass and amber beads which were worn in different compositions around the neck, on the chest, in the hair, and around the wrists. There is a clear typological development in the composition of beads. In the earlier part of the Late Roman Iron Age there is great variation in colour and shape in contradiction to the later part of the Late Roman period, where the beads are simpler and uniform. This development is not special for Fig. 11. Necklace of amber- and glass beads from Kraghave Ødetofter grave 823 (phase C3, woman age 40) (photo: Kroppedal Museum)

53 Glass and Amber Beads in Late Roman Iron Age. Relations between Denmark and the Black Sea Area with a Special Focus on the Graves from Eastern Zealand 53

54 54 Linda Boye, Ulla Lund Hansen Fig, 12. Necklace and armlet of amber and glass beads from Kong Svends Park grave 1001 (phase C1b C2, child) (photo: Kroppedal Museum)

55 Glass and Amber Beads in Late Roman Iron Age. Relations between Denmark and the Black Sea Area with a Special Focus on the Graves from Eastern Zealand 55 Fig. 13. Necklace of glass- and amber beads from Højbakkegård grave 9 (phase D, woman age 20 40) (photo: Kroppedal Museum)

56 56 Linda Boye, Ulla Lund Hansen Fig. 14. Various necklaces from female graves from the grave field Ellekilde (photo: Line Thorup, Kroppedal Museum) the eastern part of Zealand, but it is especially clear in this very area because of the presence of the many burial sites. In the area of studies, West of Copenhagen, 86 graves dated to the Late Roman Iron Age C1b C3 were excavated until of the graves belonged to males, 29 to females, and 38 graves are not sex-linked. However, it is certain that many of the 38 burials belonged to children. Out of the 19 male graves three contained one to three beads, for example the Brøndsager princeps was buried with a necklace consisting of three beads, a gold coin and three bucket-shaped bronze-pendants. 26 of the female burials contained beads. Out of the 38 indeterminable graves six contained beads, and these did all belong to children. So one must conclude that men now and then wore few beads and amulets, which also is documented from the Late Roman Iron Age weapon deposits. Generally the composition of necklaces was symmetrical and it was arranged around the largest or a different looking central placed bead. Often larger beads also were used to divide necklaces in evenly spaced segments. It was a custom to wear a necklace consisting of small beads, most often of amber, tight around the neck, and the composition of necklaces with glassand amber beads and their way of wear appear to be reasonably consistent for the whole of Zealand. Parallels are found on Funen (Henriksen 2009: 135 ff.), on Bornholm (Rasmussen 2010: 175 ff.) and in South Sweden (Stjernquist, Beck, Bergström 1994: 8 ff.). A greater variation is seen in Jutland. South of Scandinavia, in Poland and farther south and east at the Continent the bead necklaces are mostly composed and worn like in Scandinavia apart from Jutland (Mączyńska 1997: 103 ff.; Cieśliński 2009: 193 ff.; Khrapunov 2012: 183 ff.; Stoyanova 2013: 128 ff.). In 2009 Per Ethelberg summarized how woman wore their brooches during the Late Roman Iron Age. Furthermore he underlined the amount of brooches, which belonged to the high status female dress (Ethelberg 2009: 32 Abb. 17). Now one can to this add the number and size of necklaces in female graves as an extra indicator of status (fig. 16), and furthermore that from the Late Roman Iron Age C1b and C2 to C3 also great changes occurred in

57 Glass and Amber Beads in Late Roman Iron Age. Relations between Denmark and the Black Sea Area with a Special Focus on the Graves from Eastern Zealand 57 Fig. 15. Necklace of berloque-shaped amber beads from Snubbekorsgård grave 260 (photo: Erling Pultera, Kroppedal Museum) the composition of necklaces. The choice of beadtypes for constructing the necklaces and the number of beads incorporated also clear illustrate status. During period C1 C2 beads are found in many of the female graves and usually the bead strings from this period have great variation in colour and type of beads never mind each string is strictly symmetrical constructed (fig. 17). During period C3 the number of graves with beads is far smaller than in the previous periods. In period C3 however the graves contained beads in a much larger quantity, but the necklaces are quite different from those from C2. Furthermore the beads are more uniform in as well colour as shape and maybe also less complicated to produce. Due to the high degree of uniformity in C3 the diversity of the beads strings is limited. In C3 the amount of beads are dominated by brick red and yellow beads some with few yellow or red small spots, which separates them from the other brick red and yellow beads. One can observe these very seldom identical beads at the Ellekilde grave field (fig. 18). The same is the situation at the Stålmosegård / Vindinge grave field (close to the town Roskilde) (Høj 2011: ), just now being analysed by the authors. Amber beads Amber is just as glass an important material for bead production, and amber does occur in archaeological contexts in most parts of Europe (fig. 19). The origin of amber as raw material for beads has been discussed, but modern analyses hold near all European amber for having its origin

58 58 Linda Boye, Ulla Lund Hansen Fig. 16. Illustration showing the documented wear of two, three or four fibulae in female graves from phase C1b to D (Wealth and Prestige 2009: 32, Abb. 17; drawing: J. Andersen)

59 Glass and Amber Beads in Late Roman Iron Age. Relations between Denmark and the Black Sea Area with a Special Focus on the Graves from Eastern Zealand 59 Fig. 17. Necklace of amber and glass beads from Engbjerg grave 12 (phase C2, girl aged six to eight years) (photo: Kroppedal Museum)

60 60 Linda Boye, Ulla Lund Hansen Fig. 18. Opaque glass beads with red or yellow dots from Ellekilde grave 37 (phase C3, remains from two individuals, aged years) (photo: Line Thorup, Kroppedal Museum) from the Baltic Sea area and Denmark. It is not possible chemically to distinguish between amber from Denmark and the Baltic Sea as it belongs to the same formation (Heck 1997: 28 ff.). Export of amber from the North has been important during all times and do tell about far reaching exchange systems emanating from the northernmost Europe (Wielowiejski 1996: 217 ff.; 1997: 95 ff.; Bitner- Wróblewska 2010: 141 ff.). During Early Roman Iron Age trade with amber has a southerly destination towards among others Carnuntum at the Limes and from there to Aquileia, which was the early centre for manufacturing amber objects. From the middle of the second century AD the interest in amber in Italy and the Central part of the Roman empire decreased, but now the interest rose in the western Roman provinces and in East Europe. This lasted until the fifth century AD. Amber beads and amber objects are excavated in rich graves in the economic important western Roman provinces, and

61 Glass and Amber Beads in Late Roman Iron Age. Relations between Denmark and the Black Sea Area with a Special Focus on the Graves from Eastern Zealand 61 Fig. 19. Small necklace of exclusively amber beads from Engbjerg grave 4 (phase C2, woman age 20 40) (photo: Erling Pultera, Kroppedal Museum) here near the Germanic territory the fashion of berloque-shaped amber beads is copied, as illustrated for example in rich aristocratic female graves as Weiden at Cologne and in Niederzier-Hambach. The distribution show that the more south from Cologne and the Rhine region one get, the rarer is the amber (Lund Hansen 1991: 182 ff.). The limited number of berloque-shaped pendants in West and Central Europe are also known from Germanic aristocratic graves belonging in the West to early Frankish graves (Böhme 1974), in Central Germany to the so called Hassleben-Leuna group and more south also in Alamannic graves as for example Gerlachsheim and Gundelsheim (Christlein 1979: Taf. 30, 35). The distribution maps by Lund Hansen (1995: 219, fig. 8. 8), Magdalena Tempelmann-Mączyńska (1985: Taf. 70) and Mark Shchukin (Щукин 1977: 85, рис. 4, 8) demonstrate a main distribution of amber beads, including the berloque-shaped amber beads, in South Scandinavia, Poland, the southern Baltic Sea area and in minor amounts in Middle Europe and all the way south-eastwards in the Chernyakhov culture to the Black Sea region and even further away (Мастыкова 1999: 176 f.; Magomedov 2011: 179 ff.). The numerous simple disc-shaped amber beads are difficult to date without context. In Scandinavia this kind of bead is well known from the Late Roman Iron Age and Migration period. In contradiction to the simple amber beads the berloqueshaped amber beads have a characteristic distribution pattern, a limited dating and an occurrence in graves through which they tell about social systems and exchange relations. Interesting is, that the earliest appearance of the berloque-shaped amber beads in Germania Libera is in Denmark including Bornholm and in the northernmost part of Poland in phase C1b. Here they are that early found in great amounts the

62 62 Linda Boye, Ulla Lund Hansen Fig. 20. Necklace of berloque-shaped amber beads from Ellekilde grave 120 (phase C3, double grave with two individuals of unknown sex, age 45 and 35 45) (photo: Line Thorup, Kroppedal Museum) invention area. The use and wear is related to the higher social groups. Apparently not all in society had legal access to amber, and it looks like there must have been regulations concerning which part of the society were allowed to wear amber and to trade in amber (Lund Hansen 1991; 1995: 217 ff., 223, fig. 8. 9). The preference of amber beads and maybe especially the berloque-shaped amber beads in the eastern part of Central and Southeast Europe must in the very beginning depend on its origin: 1) beads of an exotic material, 2) the special berloque shape, and 3) symbolism of social status (Lund Hansen 1991; 1995: 217 ff; Мастыкова 1999) (fig. 20). It is interesting to see the striking coincidence of spread comparing the berloque-shaped amber beads and the distribution of Germanic Late Roman Iron Age iron combs (Levada 2000), of small rosette fibulae (Ethelberg 2000), of rosette fibulae (Lund Hansen, Przybyła 2010: 241 ff.), of cut glass beakers of type Kowalk (Gavritukhin 2011: 42, fig. 2) compared with the distribution maps by Magomedov (2011: 177, fig. 1) and Igor Gavritukhin (2011: 42, fig. 2) concerning glass beakers of type Eggers 230 (type Kowalk) and the distribution of the thick walled cut-glass beakers by Gavritukhin (2011: Fig. 21. Amber and glass beads from several grave finds in the area west of Copenhagen (photo: Erling Pultera, Kroppedal Museum)

63 Glass and Amber Beads in Late Roman Iron Age. Relations between Denmark and the Black Sea Area with a Special Focus on the Graves from Eastern Zealand 63

64 64 Linda Boye, Ulla Lund Hansen 52, fig. 6) and finally adding to this the map of a special type of pendants and brooches (Quast 2011: 202, fig. 5; 204, fig. 6) then it is obvious, that one deals with contacts across the European Continent in Late Roman time. Apart from berloque-shaped amber beads there are several other signs of contact between South Scandinavia and the regions south of the Baltic Sea coast not only fibulae and other jewellery as for example the peculiar neck-ring from Kong Svends Park grave 1060 (Cieśliński 2009: 193 ff.). Another example is the remarkable distribution pattern of a limited number of Roman imports as for example the glasses of type Eggers 189 (Kokowski 2001), similarities between some aristocratic very early Late Roman Iron Age graves at Zealand (for example Himlingøje grave 1980) and the North Polish Late Roman grave field Wecklice (grave 208) (Natuniewicz-Sekuła, Okulicz-Kozaryn 2008) and Pruszcz Gdański (Pietrzak 1997). The above discussed material makes it seem obvious, that there were relations between South Scandinavia and the Black Sea region. The objects do illustrate contact between aristocracies in Late Roman time concerning jewellery fashion and a bead fashion linked to a special raw material as the amber. In the beginning of the Late Roman time the berloque-shaped pendant had a limited use and distribution its first appearance is at a social elite around the Baltic Sea, and from there this special bead-fashion over time spread to the whole of Europe supported by people carrying out trade relations and now and then even underlining and securing these connections by marriages between families inside this vast geographic area of Europe. How all this in detail was carried out that is interesting to think about but also difficult to answer. Some of these questions will hopefully be solved in the next few years, when new natural scientific methods will allow us to examine the components of the individual beads (Fig. 21). Conclusion Despite the glass beads themselves and the way of combining and wearing them vary from region to region, in phase C1 and C2 they were probably produced outside Germanic area but still unknown where. It is also a question whether the change in bead preference sometimes around C3 does mean, that there were established Germanic workshops producing less complex beads, or one has to count with a shift of trade routes. The beads from C1 and C2 must illustrate different wishes and opportunities in glass bead fashion. It seem that there in some regions of Europe as Denmark, Central and Eastern Europe and Crimea at the same time can be observed a shift in the preference or access to glass bead types despite it not is the same type of beads. Future work will hopefully get nearer this problem through cooperation. Bibliography Bitner-Wróblewska A. North-Eastern Poland in First Centuries AD a World Apart // Worlds Apart? Contacts across the Baltic Sea in the Iron Age. Network Denmark Poland, / Det Kongelige Nordiske Oldskriftselskab, Nordiske Fortidsminder. Serie C. 7. København; Warszawa, Böhme H. W. Germanische Grabfunde des 4. bis 5. Jahrhunderts zwischen unterer Elbe und Loire. München, Boye L. Glasskår i munden en upåagtet gravskik i yngre romersk jernalder // Drik og du vil lev skønt. Festskrift til Ulla Lund Hansen på 60-års dagen 18. august 2002 / Publications from the National Museum. Studies in Archaeology and History. 7. København, Boye L. Bosættelsesmønstre på Østsjælland // Öresund Barriär eller Bro? Kulturelle kontakter och samhällsutveckling i Skåne och på Själland under järnålderen / Centrum för Danmarksstudier. 18. Göteborg, Boye L. Catalogue of the Burial Sites // Wealth and Prestige. An Analysis of Rich Graves from Late Roman Iron Age on Eastern Zealand, Denmark / Kroppedal. Studier i Astronomi, Nyere Tid, Arkæologi. II. Kroppedal, 2009.

65 Glass and Amber Beads in Late Roman Iron Age. Relations between Denmark and the Black Sea Area with a Special Focus on the Graves from Eastern Zealand 65 Boye L. Kong Svends Park: Cemetery with Connections between the Polish Wielbark Culture and Eastern Denmark in the Late Roman Iron Age // Worlds Apart? Contacts across the Baltic Sea in the Iron Age. Network Denmark Poland, / Det Kongelige Nordiske Oldskriftselskab. Nordiske Fortidsminder. Serie C. 7. København; Warszawa, Boye L. Lots of postholes but how do we progress? // The Iron Age on Zealand Status and Perspectives / The Royal Society of Northern Antiquaries. Copenhagen, Christlein R. Dia Alamannen. Archäologie eines lebendigen Volkes. Stuttgart, Cieśliński A. Artefacts from the cemetery at Kong Svends Park with Southern Baltic connections // Wealth and Prestige. An Analysis of Rich Graves from Late Roman Iron Age on Eastern Zealand, Denmark / Kroppedal. Studier I Astronomi, Nyere Tid, Arkæologi. II. Kroppedal, Engelhardt C. Skeletgrave på Sjælland og I det østlige Danmark. En skitse fra den ældre jernalder // Aarbøger for Nordisk Oldkyndighed og Historie. København, Ethelberg P. Skovgårde. Ein Bestattungsplatz mit reichen Frauengräbern des 3. Jhs. auf Seeland / Nordiske Fortidsminder. Ser. B. 19. København, Ethelberg P. Die Fibeln // Wealth and Prestige. An Analysis of Rich Graves from Late Roman Iron Age on Eastern Zealand, Denmark / Kroppedal. Studier i Astronomi, Nyere Tid, Arkæologi. II. Kroppedal, Exploring the Cemetery of Neyzats. Simferopol; Kristiansand, Fonnesbech-Sandberg E. Torstorp Vesterby. A cemetery from the Late Roman Iron Age // Journal of Danish Archaeology. 14. København, Gavritukhin I. Cut Glass Beakers within the Context of Studies in the Connections between the South of Eastern Europe and Scandinavia in the late period of Roman Influence and the Great Migration periods // Inter Ambo Maria. Contacts between Scandinavia and the Crimea in the Roman Period. Kristiansand; Simferopol, Heck G. Analyse und Herkunftsbestimmung // Archäologie in Deutschland Henriksen M. B. Brudager Mark en romertidsgravplads nær Gudme på Sydøstfyn. I II / Fynske Jernaldergrave Fynske Studier. 22. Odense, Høj M. Stålmosegård a cemetery from the Late Roman Iron Age // The Iron Age on Zealand Status and Perspectives. Copenhagen, Iversen R. Ellekilde en gravplads fra yngre romersk jernalder med fyrstegrav og cirkusbægre // Aarbøger for Nordisk Oldkyndighed og Historie Khrapunov I. The Crimea in the Early Iron Age. An Ethnic History. Simferopol; Kristiansand, Levada M. Metal combs of the second quarter of the first millennium AD in Eastern Europe // Die spätrömische Kaiserzeit und die frühe Völkerwanderungszeit in Mittel- und Osteuropa. Łódź, Lund Hansen U. Berlockförmige Bernsteinperlen die europäischen Kontakte eines ostdänischen Zentrums der jüngeren Kaiserzeit // Regions and Reflections. In honour of Märta Strömberg / Acta Archaeologica Lundensia. Ser. 8. No. 20. Lund, Lund Hansen U. et alii. Himlingøje Seeland Europa. Ein Gräberfeld der jüngeren Kaiserzeit auf Seeland, seine Bedeutung und internationalen Beziehungen / Nordiske Fortidsminder. Ser. B. 13). København, Lund Hansen U. Varpelev // Reallexikon der Germanischen Altertumskunde. Bd. 32. Berlin; New York, 2006a. Lund Hansen U. Glass and Beads // Sorte Muld, Wealth, Power and Religion at an Iron Age central Settlement on Bornholm. Rønne, 2009b. Lund Hansen U. Analyse von Perlen // Wealth and Prestige. An Analysis of Rich Graves from Late Roman Iron Age on Eastern Zealand, Denmark / Kroppedal. Studier i Astronomi, Nyere Tid, Arkæologi. II. Kroppedal, 2009c.

66 66 Linda Boye, Ulla Lund Hansen Lund Hansen U., Przybyła M. J. Rosettenfibeln ein Klassifikationsversuch // Worlds Apart? Contacts across the Baltic Sea in the Iron Age. Network Denmark Poland / Det Kongelige Nordiske Oldskriftselskab. Nordiske Fortidsminder. Serie C. 7. København; Warszawa, Mączyńska M. Die barbarische Kette der Römischen Kaiserzeit ihre Zusammensetzung am Beispiel der Černjachov-Kultur // Perlen. Archäologie, Techniken, Analysen / Akten des Internationalen Perlensymposiums in Mannheim vom 11. bis 14. November Bonn, Magomedov B. Zur Bedeutung sarmatischer Kulturelemente in der Černjachov-Kultur // Kontakt- Kooperation-Konflikt. Germanen und Sarmaten zwischen dem 1. und dem 4. Jahrhundert nach Christus / Internationales Kolloquium des Vorgeschichtlichen Seminars der Philipps-Universität Marburg, Februar Neumünster, Magomedov B. The Chernyakhov People s Contacts with Scandinavia and the Crimea // Inter Ambo Maria. Contacts between Scandinavia and the Crimea in the Roman Period. Kristiansand; Simferopol, Natuniewicz-Sekuła M., Okulicz-Kozaryn J. Two richly furnished graves with Roman imports from the cemetery at Wecklice, site 7, Elbląg commune (Poland) // Germania. 86. Frankfurt, Pietrzak M. Pruszcz Gdański Fundstelle 10. Ein Gräberfeld der Oksywie- und Wielbark-Kultur in Ostpommern / Monumenta Archaeologica Barbarica. IV. Kraków, Rasmussen B. M. Slusegårdgravpladsen V. Fundoversigt og genstandstyper // Jysk Arkæologisk Selskabs Skrifter. XIV/5. Århus, Stjernquist B., Beck C. W., Bergström J. Archaeological and scientific studies of amber from the Swedish Iron Age // Scripta Minora. Studier udgivna av Kungl. Humanistiska Vetenskabssamfundet i Lund. 1. ( ) Lund, Stoyanova A. The Second and the First Half of the Third Century AD Female Costume Accessories in the Cemetery of Neyzats // Exploring the Cemetery of Neyzats. Simferopol; Kristiansand, Tempelmann-Maczyńska M. Die Perlen der römischen Kaiserzeit und der frühen Phase der Völkerwanderungszeit im mitteleuropäischen Barbaricum / Römisch-germanische Forschungen. 43. Mainz, The Iron Age on Zealand Status and Perspectives / The Royal Society of Northern Antiquaries. Copenhagen, Wealth and Prestige. An Analysis of Rich Graves from Late Roman Iron Age on Eastern Zealand, Denmark / Kroppedal. Studier I Astronomi, Nyere Tid, Arkæologi. II. Kroppedal, Wielowiejski P. Bernstein in der Przeworsk-Kultur // BRGK Wielowiejski P. Bernsteinperlen und römisch-barbarische Kulturkontakte in Mitteleuropa // Perlen. Archäologie, Techniken, Analysen / Akten des Internationalen Perlensymposiums in Mannheim vom 11. bis 14. November Bonn, Quast D. The Links between the Crimea and Scandinavia: some jewellery from the third century AD princely graves in an international context // Inter Ambo Maria. Contacts between Scandinavia and the Crimea in the Roman Period. Kristiansand; Simferopol, Мастыкова А. В. О распространении янтарных грибовидных бус-подвесок позднеримского времени на юге Восточной Европы и в Закавказье // 100 лет черняховской культуре. Киев, Щукин М. Б. Современное состояние готской проблемы и черняховская культура // Археологический сборник

67 Glass and Amber Beads in Late Roman Iron Age. Relations between Denmark and the Black Sea Area with a Special Focus on the Graves from Eastern Zealand 67 Линда БОЙЕ, Улла ЛУНН-ХАНСЕН Стеклянные и янтарные бусы позднеримского времени между Данией и Причерноморьем с акцентом на погребения в восточной Зеландии Резюме Существуют находки, в особенности многочисленные стеклянные сосуды для питья, которые очевидно указывают на контакты между Скандинавией и Юго-Восточной Европой в позднеримское время и эпоху Великого переселения народов. К показателям таких контактов между отдалёнными регионами следует добавить и бусы. В североевропейских погребениях ранне римского времени (фазы В1 В2) бусы встре чаются редко, а если они время от времени появляются, то всегда в богатых женских могилах с многочисленным инвентарём. С началом позднеримского времени бусы часто встречаются в погребениях женщин, а также на «центральных поселениях» (торговых центрах). На протяжении большей части позднеримского времени бус было много и они были самые разные. В богатых женских погребениях очень часто были уникальные наборы бус. Приблизительно в начале фазы С3 изменяется материал, из которого делали стеклянные бусы возможно, в связи с модой на бусы, возможно, из-за доступности стеклянных бус как в южной Скандинавии, так и в варварской Европе в целом. В южной Скандинавии воз растает среднее количество бус в могиле и од новременно изменяется тип стеклянных бус, причём набор типов стеклянных бус становится менее разнообразным. На протяжении трёх последних десятилетий исследования в Западном Копенгагене выявили девять погребальных памятников, датированных позднеримским временем (фазы C1b C3) и начала эпохи Переселения народов (фаза D). Помимо полей погребений, раскопаны поселения с тремя типами «длинных домов» «Хёйе Тоструп», «Рагнесминде» и «Стувехёй». Каждый из этих типов можно связать с определенным классом или прослойкой позднеримского общества. Практически во всех женских погребениях в этом регионе находят стеклянные или янтарные бусы, в разных сочетаниях на шее, на груди, в волосах и вокруг кистей рук. Наблюдается явная типологическая эволюция состава этих бус. В начале позднеримского времени расцветка и форма бус была чрезвычайно разнообразной в противоположность концу позднеримского времени, когда бусы стали проще и однообразнее. Эта эволюция присуща не только восточной части Зеландии, но в этом регионе её можно проследить более отчётливо, поскольку тут имеется много погребальных памятников. Кроме того, бусы найдены во многих детских могилах, а также в нескольких мужских погребениях последние сопровождались всего одной-тремя бусинами. Существовала традиция носить ожерелье из мелких бусин, чаще всего янтарных, тесно обвивавшее шею, а состав ожерелий из стеклянных и янтарных бус, а также способ их носки, как выясняется, были достаточно близки на всей территории Зеландии. Помимо этого, в позднеримское время обычным материалом для изготовления бус стал янтарь. В начале позднеримского времени использование и распространение грушевидных подвесок-кулонов было ограниченным. Впервые они появляются у социальных элит побережья Балтийского моря (в Дании, южной Скании, Прибалтике и северной Польше), а оттуда мода на этот своеобразный тип бус со временем распространяется по всей Европе благодаря людям, которые вели торговлю и эпизодически создавали и поддерживали родственные связи. Предпочтение, которое на востоке Центральной и в Юго-Восточной Европе оказывали янтарным бусам и особенно янтарным грушевидным подвескам-кулонам, должно было, по крайней мере изначально, объясняться их происхождением: 1) это бусы из экзотического материала, 2) особой грушевидной формы, 3) символизирующие положение в обществе. Несмотря на то, что и сами стеклянные бусы, и способ их сочетания и ношения отличаются в зависимости от региона, вероятно, в фазах С1 и

68 68 Linda Boye, Ulla Lund Hansen С2 их делали за пределами области, населённой германцам, пусть пока и неизвестно, где именно. Также остается невыясненным, насколько изменившиеся предпочтения по отношению к бусам приблизительно в фазе С3 связаны с появлением германских мастерских, где делали менее сложные бусы, или же в результатом изменения торговых путей. Бусы фаз С1 С2 должны указывать на разные вкусы и возможности, проявлявшиеся в моде на стеклянные бусы. Кажется, в некоторых европейских регионах, например, Дании, Центральной и Восточной Европе, а также в Крыму, одновременно фиксируются изменения в предпочтениях относительно типов стеклянных бус, или же в их доступности, несмотря на то, что это не одни и те же типы. Будем надеяться, что дальнейшие исследования в сотрудничестве с коллегами помогут приблизиться к решению данной проблемы.

69 69 Katarzyna CZARNECKA Padlocks in the Przeworsk and the Chernyakhov Cultures in the Late Roman Period, as an Evidence of Mutual Contacts Padlocks are portable locks used to defend against theft and unauthorized use. They are designed to protect against some degree of forced and surreptitious entry. Padlocks could be carried and used where necessary. The first evidence of use of such type of lock comes from the Celts. Earliest finds of padlocks are known from the middle La Tène period (fig. 1. 1). Cylindrical iron box with attached rings (manacles) was closed by barbed bold, inserted through a small opening, and served as a locking mechanism of shackles and fetters (Dechelette 1913: fig. 29; Nothdurfter 1979: 91, fig ; Paulik, Tomčíkova 2005: 104). Such padlocks, type 01 according to Jean Paul Guillaumet and Gerard Laude (2009: fig. 6), were adopted by Romans together with many other Celtic technical inventions and were in use till the end of Roman empire (fig. 1. 2, 3). Beside them Roman locksmiths developed other types of padlocks: box padlocks. Some of them was closed with separated barbed bolt (fig. 1. 4), other with spring attached inside and opened with rotated key 1 (fig. 1. 5). Roman padlocks have no exhaustive description or complete classification. It is still a category of small finds which hasn t caught an attention of most of the scholars. Iron padlocks found outside the Limes have been treated as Roman imports (Zotz 1935: 25). In fact only very few artefacts of this kind found in the Barbaricum came from the Roman workshops. There is one from deposit from Burgellern, Reisberg, Lkr. Bamberg (Haberstroh 2000: pl ), one from settlement Brahlstorf, Kreis Ludwigslust in Mecklemburg (Brandt 2005: 354, fig , 3) and one, found together with a ring-key, from grave 17 on the Bogaczewo culture cemetery in Mojtyny (Moythienen), district Mrągowo in Masuria (Hollack, Peiser 1904: 45, pl. III). Padlocks were found in graves XI.2 and VII.1 on the Sarmatian/ German cemetery in Hortobágy-Poroshát, kom. Hajdu-Bihar (Zoltai 1941: 296, 303, pl. XI. 26; XIV. 7, 8). All other artefacts of that kind found outside the Limes were locally made. The very idea of padlock, technical solution, was adapted from the Roman specimens, but their form and details of construction are different in padlocks known from Roman provinces, and from the Barbaricum. Roman padlocks consisted of two elements: iron (rarely bronze) elongated case, most often with rectangular section, 2 with attached to the side of it a long, U-shaped bent, rod. Loose end of this rod, which was going parallel to the side, protruded a little (fig. 2. 1b). In another variant it is bent above the padlock case and ending was provided with a loop (fig. 2. 1a). Locking bolt consisted of a rod; on one end of it were attached flat bands of metal, forming springing barbs. Bolt could have been single (that is with barbs set on both side of the central rod) or double (that is with split rod with two sets of both side barbs). The other ending was bent and provided with a loop. Barbed ending was inserted into a case by a small rectangular opening in a lid; loop was put on a loose end of the parallel rod. (fig. 2. 1b). In the other variant widened ending of the bolt closed the opening on the ending of bent parallel rod. (fig. 2.1a) Springing barbs were blocked on the lid, and the padlock was closed. It was a latch lock; no key was needed to close it, only to open. A key, in a form of a small rectangular frame set at an almost right angle on a elongated handle, was inserted by a narrow slit in the bottom of the padlock case, the frame was put on the end of the bolt, and, by moving up, it pressed the barbs to the central rod. The latch was then re- 1 A very instructive reconstruction of such padlock gives J. W. Anstee (2001). 2 Padlocks dated later were sometimes cylindrical.

70 70 Katarzyna CZARNECKA Fig. 1. Celtic padlock: 1 Sanzeno, region Trentino-Alto Adige, Italy (Nothdurfter 1979: fig ); Roman padlocks: 2, 3 Vindolanda/Chesterholm, Northumberland, Great Britain (Birley 1997: fig , 84); 4 Mainz-Weisenau, Germany (Kessler 1932: fig. 5. 8); 5 Neupotz, Landkreis Germerheim, Germany (Künzl 1993: pl. 47. NJ5)

71 Padlocks in the Przeworsk and the Chernyakhov Cultures in the Late Roman Period, as an Evidence of Mutual Contacts 71 Fig. 2. 1a, b Roman padlock schemes (Manning 1985: fig ); 2 Barbarian padlock scheme (drawing: the author) leased, bolt could be removed, the padlock opened. Locks with double barbed bolt needed keys with divided frame. In some variants key hole in a shape of T, was on the side wall (fig. 1. 4). In that case key-frame was inserted in the perpendicular, shorter opening and set on the end of a bolt. Longer slit allowed moving the rod up, and pressing the barbs. The construction of padlocks found outside the Limes was similar, but there were differences in shape and technical details (fig. 2. 2). These padlocks consisted of metal cylindrical case (exclusively circular section) with alongside attached a narrow metal sleeve. Terminals were closed with flat metal lids covering both: cylinder and sleeve, provided with openings for a bolt in a lid, and for a key, in the bottom. The separate, not attached to the box, barbed bolt was bended in U shape to form a bow. Loose end of the bow was caught in a sleeve

72 72 Katarzyna CZARNECKA and any items (loops, chain-links) which were put on the bow were safely closed. A key was inserted by narrow slit in the bottom, then the frame was put on the bolt to catch the barbs, what allowed opening of the padlock in the same way as in the Roman specimens. The differences between specimens found inside the Roman provinces and these found outside the limes are not great, however clear territorial division: one type is known almost exclusively from the Roman empire, the other is known exclusively from the Barbaricum, is very significant and led to conclusion that the latter were locally made, not imported. The know-how, the technical solution, is undoubtedly Roman. The technique is not very complicated and required no special tools. It was within reach of barbarian (Germanic) smiths. What is interesting only one type (with barbed bolt) of Roman padlocks was adapted. Most probably it was easiest to produce. All padlocks known from the Barbaricum have similar form and similar dimensions: a cylindrical box circa 5 6 cm long, diameter cm. Only a few were preserved in more or less complete form, many are badly damaged. Locking bolts could be single or double (with two sets of barbs). In double bolts barbs could be set parallel or perpendicular. The later was most probably more difficult to open without the proper key. Barbs were soldered or nailed 3 to the central rod. Most often it is impossible to see how it was made, due to the state of preservation corrosion and damages. Above the barbs was usually set a small protrusion to protect from falling to deep into the padlock s case. 4 The blocking bolt is separate, not attached in any way to the padlock body and should have been somehow protected from loosing maybe by piece of string or thong. Almost all of these padlocks were made of iron. Bronze padlocks are known only from the Migration рeriod, e. g. from well-known settlement in Jakuszowice, district Kazimierza Wielka (Godłowski 1992: fig. 7. 2). 5 Keys found in the Barbaricum are just the same as specimens known from the Roman empire. They were made from the flat rod ended with perpendicular frame (fig. 3. 5, 6). In case of loose finds it is hard or rather impossible to distinguish imports from local products. More then twenty padlocks and at least ten keys were found on the Przeworsk culture territory. It is possible, that in fact there were more specimens in use, but state of preservation (e. g. only uncharacteristic corroded fragments survived) prevents from proper identification. All of them are dated to the late Roman Period and the Migration period, phases C2 D, however only in a few cases context allowed more precise dating. Padlocks were rare in graves furniture. From the Przeworsk culture we know only a few funeral finds. Quite unique is a cemetery Żerniki Wielkie (Gross Sürding), district Wrocław, where three padlocks and a key were found in graves (Zotz 1935: 65, 81, 82, figs. 6. 5; ; 23. 7; 24. 1). A padlock from grave 9 was deposited without closing bolt (it means it was open ), it was 5 cm long, diameter 2.5 cm. The specimen is damaged, a parallel sleeve is lacking. It was an inhumation grave of an adult man, furnished with three arrow heads, a knife, four buckles, a beak-shaped strap end, a comb with bellshaped back, and a big glass bead. The grave must be dated to the early Migration period. The padlock and arrowheads were placed by knees of the body. Second padlock, also without closing bolt, was found in grave 46. It is smaller, length about 4 cm, diameter 1.8 cm (fig. 3. 2). A man and woman were buried in the grave, both skeletons are badly preserved. The padlock was placed by the neck of the (most probably) female skeleton (fig. 6). Third padlock, found in grave 49, is a little bit different. A parallel sleeve is set apart from the cylindrical case joined with three separate links. A specimen is rather small, only 3.7 cm long, diameter 1.6 cm. A bolt is inside, so the padlock was deposited closed. It was placed by the arm of a skeleton of a woman, adultus. The grave was furnished with an iron brooch, type A.158. The only key was found in grave 47. All graves were, most probably dated to the end of the Roman period or early Migration period. Well preserved small padlock, similar to the find from grave 49 that is with a sleeve set apart 3 Like in dated to the Migration period specimen from Jakuszowice, district Kazimierza Wielka (Godłowski 1992: fig. 7. 2). 4 E. g. Chabielice (Frąsiak, Gwóźdź, Siciński 2000: fig ) or Piwonice, from hut 4 (Dąbrowski 1956: pl. LXXXIV. 3). 5 Bronze padlocks were more frequent on Merovingian period sites, e. g. Wörmlitz, Stadtkreis Halle (Schmidt 1975: pl e) or Weingarten Kreis Ravensburg (Roth, Theune 1995: 124, pl C. 5).

73 Padlocks in the Przeworsk and the Chernyakhov Cultures in the Late Roman Period, as an Evidence of Mutual Contacts 73 Fig. 3. The Przeworsk culture padlocks elements. 1 Zofipole, district Kraków (Jażdżewski 1976; fig. 1. 7); 2 Żerniki / Gross Sürding, grave 46 (Zotz 1935: fig ); 3 Przywóz, district Wieluń; 4, 7 Walków Kurnica, district Wieluń; 5 Mierzanowice, district Opatów (Miśkiewicz 1961: pl. XIII. 9); 6 Toporów, district Wieluń (Łaszczewska 1975: Pl. II. 1); 8 Izdebno Kościelne, district Grodzisk; 9 Piwonice, district Kalisz; 10 Chabielice, district Bełchatów (Frąsiak, Gwóźdź, Siciński 2000: fig ). All in Poland. (Drawings: 3, 4, 7 the author; 8 G. Nowakowska) and a decorative zigzag, was found, as a loose find, on a cemetery in Witaszewice, district Łęczyca (Jażdżewski 1976: fig. 2. 6) (fig. 7. 4). 6 A complete padlock, closed (that is with inserted inside bold) was found in grave 5 from a cemetery in Karczyn, district Inowrocław (Bednarczyk, Romańska 2004). It is an inhumation grave, burial of 40 years old man, furnished with iron tendril brooch, small bronze hook-pin, two iron rings, an iron handle and a knife. The padlock is 6.8 cm long, with a diameter of 3 cm. Grave was dated to the phase C3 D. Beside these few examples from cemeteries padlocks are known mostly from settlements. A padlock was found in a kiln no 13, on a pottery production settlement in Zofipole, district Kraków, dated to the phase C1b C2 (Dobrzańska 2000: 48, fig. 12.1) (fig. 3. 1). In Igołomia, district Kraków, 6 I would like to thank Dr. Wojciech Siciński from the Museum Archeologiczne i Etnograficzne in Łódź for the kind permission to publish a photo of this object.

74 74 Katarzyna CZARNECKA also a pottery production settlement, a damaged cylindrical case was found in a dwelling object (a hut), together with a brooch type A.158, a buckle, a silver coin (denarius) of Hadrianus, spindle whorls, loom weights, beads, needle, hook pin, crucibles, and many artefacts made of bone, antler, slag, and potsherds (Dobrzańska 1990: 38, pl. XXVIII. 7). Such rich set of various finds may indicate that this building served as a workshop. The object can be dated to the Late Roman period. A fragment of a blocking bolt was found (as a loose find) on another settlement with pottery workshops, in Chabielice, district Bełchatów (Frąsiak, Gwóźdź, Siciński 2000: 77, fig ) (fig ). Parts of at least three padlocks were found on the settlement Piwonice, district Kalisz, site 1. A blocking bolt was in a hut, together with a belt buckle of Strzegocice type, so the object can be dated to phase D2 (Dąbrowski 1956: 185, pl. LXXXIV. 3) (fig. 7. 3). Another bolt and a padlock case were loose finds (Dąbrowski 1958: 83, pl. XLII. 3) (fig. 3. 9). From the Piwonice site 3, a damaged key (with broken frame) was found in a house 11, dated to phase C2 (Dąbrowski, Kozłowska 1959: 323, Pl. XIX. 2). Similar chronology could have a well-preserved key found in a pit on settlement in Mierzanowice, district Opatów (Miśkiewicz 1961: 273, pl. XIII. 9) (fig. 3. 5; 7. 2). Huge rescue excavations in Aleksandrowice, district Kraków, brought at least five keys. 7 Fragment of double blocking bold was found, as a loose find, on the settlement Izdebno Kościelne, district Grodzisk Mazowiecki 8 (fig. 3. 8; 7. 1). Small corroded fragment was in Zagórzyce, district Kazimierza Wielka (Grygiel, Pikulski, Trojan 2009: 278, pl. I. 9). Well preserved padlock-case was found, unfortunately as a loose find, in the rich settlement of Przywóz, district Wieluń (Makiewicz 1969: 101, fig. 4; Jadczyk 1971: 178, fig. 5c) (fig. 3. 3). Two openings in a lid suggest that it was closed with a double bolt. An iron key (fig. 3. 6) and a double bolt are known from the nearby settlements in Toporów, district Wieluń (Łaszczewska 1975: 297, pl. II. 1). Quite unique is a settlement from Walków Kurnica, district Wieluń, where at least six padlocks were used in almost the same time. Five well preserved padlock cases, one fragment of such case, two blocking bolts, and two keys were found separately in various dwelling objects, pits and as loose finds (fig. 3. 4, 7). The settlement is dated to the final stage of the Przeworsk culture. 9 There are no many finds of padlocks outside the Przeworsk culture. One was found in the West Balts area, on a cemetery of the Dolkeim-Kovrovo culture in Beryozovka (Gross Sausgarten), Bagrationovsk district (Stadie 1914: ; Archeologiczne dziedzictwo 2011: ). Grave 1 furnished with short sword (or long knife), so called Dolchmesser, two spearheads and a shield grip, and damaged silver brooch is dated to phase C3. A well preserved padlock case is 7 cm long and 3 cm in diameter with attached sleeve, endings closed with tear-shaped plates. The padlock is closed, that is a bolt is inside. More padlocks, of slightly different type, dated to the late Migration period and later, are more common in this area, but from the Roman period there is only imported specimen from Moythienen, Bogaczewo culture, and mentioned above from Gross Sausgarten. No padlocks are known from the Wielbark culture. It is quite possible that the lack of those finds in graves is due to the specific burial custom, not allowing depositing in graves iron objects like weaponry or tools. The lack of finds from settlements can be, to some extend, a result of state of researches. Not many Wielbark culture settlements were excavated. Padlocks very similar in shape and dimensions to artefacts known from the Przeworsk culture are quite numerous on sites of the Chernyakhov culture, both cemeteries and settlements. Iron padlock comes from the inhumation grave 20 from the cemetery in Kaborga IV, Nikolayev oblast, Ukraine (Магомедов 1979: 44, 46, таб. XIII. 5, 7; 1987, fig ). (fig. 4:1) An iron case is well preserved, 5 cm length, diameter 2.7 cm. A bolt is preserved in two fragments. Among the grave furnishing was also a flat U-shaped specimen, which could be a part of broken key frame. A grave with a bell shaped comb could be dated to phase C3 D. Damaged, iron, cylindrical padlock case was found in a burial of a child in grave 31 from cemetery in Gavrilovka, Kherson oblast, furnished with a brooch with trapezoid foot, needle, knife, a frag- 7 These finds were interpreted by Dr. Ryszard Naglik, who made this excavation, as imports from the Empire. In my opinion they could be locally made. As I mentioned before the form is so simple, that it is almost impossible to distinguish Roman and local keys. 8 Unpublished, State Archaeological Museum Warsaw, inv. number PMA-IV-9546/ There are some radiocarbon dates from the end of 5th and beginning of the 6th centuries (Abramek 2004: 330).

75 Padlocks in the Przeworsk and the Chernyakhov Cultures in the Late Roman Period, as an Evidence of Mutual Contacts 75 Fig Kaborga IV, Nikolayev oblast (Магомедов 1979: таб. XIII. 5, 7); 2 Gavrilovka, Kherson oblast (Сымонович 1960: pl. XI. 1; XIV. 7; XV. 69); 3 Repnev, Lvov oblast (Бapaн 1964: рис , 23); 4 Barbara, Poltava oblast (Башкатов 2008: рис. 8. 8); 5, 6 Bârlad-Valea Seacâ, Vaslui district (Palade 2004: fig. 3. 8; 13. 2); 7 Druzhnoye, Crimea (Храпунов 2002: 36, fig , 10). 1 4, 7 Ukraine; 5 6 Romania ment of iron ring and bucket shaped pendant (?). The two last mentioned objects could be interpreted as fragments of a blocking bolt bow and of the attached sleeve 10 (fig. 4. 2). The grave can be dated to phase C2 (Сымонович 1960: , pl. XI. 1; XIV. 7, XV. 69). Very interesting is find from grave 11 from Ruzhichanka, Khmel nitskiy oblast. In this inhumation grave, dated to the late Roman period, were found three imported Roman casket keys and a fragment of, most probably, padlock case. It is a piece of iron cylinder, diameter of 2.5 cm with 10 It is only a suggestion corroded iron pieces could be very deceptive.

76 76 Katarzyna CZARNECKA closing plate without opening for key or blocking bolt (Винокур 1972: рис ). Padlocks with side slit for key are known from the Roman empire (Gáspár, Salamon 1986: fig. 1) and the find from Ruzhichanka could have been an import there are other imports in the grave but it cannot also be excluded, that it is not a padlock fragment, but some other artefact. Padlocks are known also from the settlements. In Repnev II, Lvov oblast, an iron key was found in hut 13, together with fragments of padlock and set of carpenter tools (Бapaн 1964: , рис ) (fig. 4. 3). The object is dated to the late Roman period. Very interesting is find from settlement Barbara, in hut no. 1, Poltava oblast (Башкатов 2008: 9, 15, рис. 8. 8). According to the author, a damaged padlock laid by the broken wall dividing two rooms in the hut. A cylindrical case has 2 cm diameter, length more than of 4.5 cm. (fig. 4. 4). A whole object is dated to the Late Roman period. On the settlement Voytenki, Kharkov oblast, two blocking bolds were found, unfortunately as loose finds. 11 Two badly corroded padlocks were found on the settlement Bârlad Valea Seacâ, judeţ (district) Vaslui, in Romania. One of them, from the hut number 3, is a cylindrical case with closing plates, sleeve and bolt lacked (Palade 2004: 46, fig. 3. 8) (fig. 4. 5). In this hut, beside knife, sickle and a spindle whorl were found many pieces of antler, rough and semi-finished, most probably used to comb production. Another padlock was found in the hut 13, only cylindrical case with closing plates preserved (Palade 2004: 49, fig ) (fig. 4. 6). The specimen is 4.5 cm long with diameter of 2.5 cm. It was found together with a belt buckle of type AD30, according to Renata Madyda-Legutko (1987), a loom weight and pottery, both hand- and wheel-made. A settlement was used primarily by free Dacians, than by the Sîntana de Mureş culture, and it was a centre of comb production. More than 30 workshops were discovered there. Padlocks should be connected with the Sîntana de Mureş culture phase. A fragment of a padlock: characteristic closing plate with a slit for a key was found, unfortunately as a loose find, on the Sîntana de Mureş culture cemetery in Budeşti, Moldavia (Vornic 2006: fig ). Padlocks, however not many, are known from the Sarmatian cemeteries. Complete specimen was found in the robbed catacomb grave 85 from Druzhnoye in the Crimea (Храпунов 2002: 36, рис ). The padlock is a bit asymmetrical, length cm, diameter 3 cm, with attached sleeve, both endings covered with tear-shaped plates. In one of them is a square opening for a blocking bolt and in the other end a slit for a key. The key is badly corroded and a frame is not complete (fig. 4. 7). Grave was richly furnished with many items suggesting both man s and woman s burial, that is glass beads, silver earrings, rings, brooches and spindle whorl, but also axe and horse-bit, and tools of various kind: knives, awls, whetstones, and also pottery. Very important find are four bronze coins of Maximinus Daia, Licinius, and Galerius, which allow dating the grave to the fourth century. Similar chronology has a grave 9 from Sarmatian cemetery Drăgăneşti-Olt, judeţ. Olt, Romania (Trohani, Zorzoliu 1983: 216, fig. 5:2). In this grave was found iron padlock with bronze elements, 6 cm long, diameter 2.2 cm. Most padlocks known from the Barbaricum were found in the territory of the Przeworsk and the Chernyakhov cultures (fig. 5). They seem to be very similar in shape and dimensions. These are cylindrical cases, 5 6 cm long, cm in diameter, with a sleeve attached to the side. Unfortunately, most of the finds of padlocks from the Przeworsk culture, and some from the Chernyakhov culture are loose finds, what allows dating only in frames of the chronology of a site. 12 Dated finds, both in the Przeworsk and the Chernyakhov cultures, have the same chronology Late Roman period from phase C2 to the Early Migration period. It is hard to say if the idea of using padlocks came from the Przeworsk culture to the Chernyakhov culture or vice versa, or, what is also possible, simultaneously from various places. There are some, however not many evidences of contacts and mutual influences of these two cultures seen in adapting some elements of burial rite and also some types of pottery (Godłowski 1985: 107; Dąbrowska 1995: 105, Магомедов 2001: 34, 61). 11 I would like to thank Dr Mikhail Lyubichev for this information. 12 Additional problem emerged when these specimens were found on the surface of the site, because in such cases chronology and culture is even harder to establish. Early medieval padlocks are more or less similar (with parallel sleeve and common plates on endings) and when the padlock is fragmentary preserved and found loose on the site where also later materials were found it is very hard to determine the culture and chronology.

77 Padlocks in the Przeworsk and the Chernyakhov Cultures in the Late Roman Period, as an Evidence of Mutual Contacts 77 Fig. 5. Finds of padlocks: 1 Aleksandrowice, district Kraków; 2 Chabielice, district Bełchatów; 3 Igołomia, district Kraków; 4 Izdebno Kościelne, district Grodzisk Mazowiecki; 5 Jakuszowice, district Kazimierza Wielka; 6 Karczyn, district Inowrocław; 7 Kraków-Krzesławice, district Kraków; 8 Mierzanowice, district Opatów; 9 Piwonice 1, 3, district Kalisz; 10 Przywóz, district Wieluń; 11 Toporów, district Wieluń; 12 Walków-Kurnica, district Wieluń; 13 Witaszewice, district Łęczyca; 14 Zagórzyce, district Kazimierza Wielka; 15 Zofipole, district Kraków; 16 Żerniki Wielkie / Gross Sürding, district Wrocław; 17 Mojtyny / Moythienen, district Mrągowo; 18 Berezovka / Sausgarten, Bagrationovsk district; 19 Barbara, Poltava oblast ; 20 Gavrilovka, Kherson oblast ; 21 Kaborga IV, Nikolayev oblast ; 22 Repnev, Lvov oblast ; 23 Ruzhichanka, Kmelnitskiy oblast ; 24 Voitenki; Kharkov oblast ; 25 Bârlad Valea Seacâ, Vaslui district; 26 Budeşti, municipia Chişiniău; 27 Druzhnoye, Crimea; 28 Drăgăneşti-Olt, district Olt; 29 Hortobágy-Poroshát, kommune Hajdu-Bihar; 30 Brahlstorf, Kreiss Ludwigslust Poland; 18 Russia; 19 24, 27 Ukraine; 25, 28 Romania; 26 Moldova; 29 Hungary; 30 Germany Przeworsk culture; 17 Bogaczewo culture; 18 Dolkeim-Kovrovo culture; Chernyakhov culture; Sîntana de Mureş culture; 27, 28, 29(?) Sarmatian cultures; 30 Elbian circle

78 78 Katarzyna CZARNECKA Fig. 6. Żerniki Wielkie / Gross Sürding, district Wrocław, grave 46 (Zotz 1932: fig. 46) As I mentioned before, a padlock construction, technical principle is obviously taken from the Roman workshops. There is a question when and where it was adapted inside the Barbaricum. A possible place could be a region north of the Danubian provinces of the Empire and Dacia. From this area populated with people of mixed cultures free Dacians, the Sîntana de Mureş, and also the Przeworsk culture (the Upper Tisa group) we know quite a few examples of adapting from the provincial workshops various Roman technical abilities. First of all it is a wheel-made pottery production, sometimes with specific decoration (e. g. stamp marks). It is confirmed on many settlements located outside the limes (but close to it) from today northern Romania Transilvania (Gindele 2011: 34, 436). It is possible, that also Roman way of closing with padlocks was adapted there. An origin of the wheel made pottery workshops from the Przeworsk culture as a result of connections with Dacian culture, mostly free Dacians from Muntenia, and through them with Roman provinces, have been already seen by Kazimierz Godłowski (1982: 68). Similar opinion presents Judyta Rodzińska-Nowak (2006: 202). The earliest finds of padlocks from the Przeworsk culture come from the settlements production centres of wheelmade pottery near Krakow. Find from the kiln 13 in Zofipole is dated to phase C1b C2 (Dobrzańska 2000: 48), and from hut 70/55 from Igołomia (Dobrzańska 1990: 38) to phase C2a. That allows a cautious suggestion that beginning of using of padlocks in the Przeworsk culture may be connected with the start of organized pottery production, at the beginning of the phase C1, and was brought by the same workers, coming from the south east. The Upper Tisa group could help to establish contacts with the rest of the Przeworsk culture from Polish territory. Such possibilities discussed lately Halina Dobrzańska (2011: , fig. 21). Two Roman padlocks were found in graves on the cemetery Hortobágy-Poroshát, kommune Hajdu-Bihar in Hungary. The ethnicity of people buried there (Sarmatians? Germans?) was discussed by scholars, but all pointed to the long distance connections with the Przeworsk culture and mutual influences, seen mostly in weaponry (Istvánovits, Kulcsár 2000). It is not excluded that Roman padlocks used by these people inspire local adaptation, which was transferred to the Przeworsk culture area where it was developed and disseminated. There is a question how and from whom adopted the Chernyakhov culture the idea of using / producing padlocks. It could happened in a similar way, that is from south western neighbours free Dacians and Sîntana de Mureş culture. Boris Magomedov (Магомедов 2001: 61) sees in pottery forms and kilns types, however only from the upper and middle Dniester region, references to the Przeworsk culture. It is possible that the wheel made pottery production, at least in the mentioned area, was taken by the Chernyakhov culture workers from the Przeworsk culture centres. The Upper Tisa group could be also here a meeting point (Dobrzańska 2011: 283). What is interesting padlocks are known from the settlements with pottery production centres. A good example could be Repnev, Lvov oblast, where early kilns were found, and, in a hut, a complete padlock, and also Voytenki, Kharkov oblast, where two blocking bolts were found, as a loose finds. Unfortunately, such finds are not numerous, so presented possibility must be treated only as a hypothesis. In discussion about the possible source of wheel-made pottery production in the Chernyakhov culture are also mentioned Greek towns at the Black Sea, like Olbia (Магомедов 2001: 61). There is, however, no evidence of using padlocks with barbed blo cking bolt

79 Padlocks in the Przeworsk and the Chernyakhov Cultures in the Late Roman Period, as an Evidence of Mutual Contacts 79 Fig Izdebno Kościelne, district Grodzisk Mazowiecki; 2 Mierzanowice, district Opatów; 3 Piwonice 3, district Kalisz, hut 4; 4 Witaszewice, district Łęczyca, loose find. 1 4 Poland (photos: 1, 2 R. Sofuł; 3 A. Kędzierski; 4 W. Pohorecki) in these towns, but it can be a result of the state of researches and publication, because small finds, not always well preserved, were often omitted. The question from whom the Chernyakhov culture adapted an idea of padlock is still open. It must be mentioned that this hypothetical model: adapting Roman padlocks together with adapting wheel made pottery production, is not confirmed in other parts of the Barbaricum. Peoples of the Elbian circle, from central Germany, who learned from Romans so many various technical skills, among them the wheel-made pottery production, did not use padlocks. They know these devices, there are single imports (Brahlsdorf), but there is no local adaptation. Maybe they used locks of another type. On the rich settlement production centre in Frienstadt in Thuringia, many Roman keys were found, but no padlocks. 13 A question: why not all cultures adapted, and disseminated on own area this type of lock is still without answer. Padlocks from settlements were usually found in houses with rich furnishing, 14 sometimes with traces of metallurgical production. 15 Hut 3 from Bârlad-Valea Seacâ was a comb maker workshop, and in the hut 13 from Repnev a set of carpenter tools was found. Padlock found in Przywóz was a loose find, but the settlement is generally rich and provided many unique imports from the Roman empire. The way of use and function of padlocks seems quite obvious and unchanged till present days. However question: what exactly was closed by them, and how, exactly, they were used, is not so easy to answer. In the Roman empire padlocks were mostly used as closure of shackles and fetters. This function is not confirmed in the Barbaricum. Even evident shackle-lock from Mojtyny (Moythienen) was most probably not used in that way. Padlocks could have served to close some special rooms with maybe more massive doors, but we do not have enough information about the construction of the Roman period houses to confirm it. Padlocks could be used to close big casts or boxes. This seemed most probable. A lid and a side of such wooden cast 13 Personal communication from Dr. Christoph Schmidt, for which I would like to thank. 14 E. g. Piwonice site 3, house 11 with many various silver ornaments, including silver and glass pendants (Dąbrowski, Kozłowska 1959: 323, pl. XIX). 15 A crucible with traces of silver was found in house 70/55 in Igołomia (Dobrzańska, 1990: 38, pl. XXVIII:4).

80 80 Katarzyna CZARNECKA must be provided with loops or rings, on which the bend bow of a padlock was set (just like in modern device), and then closed. Locks elements and keys, mostly from caskets, were found almost exclusively in graves of women, and are treated as indicators of burials of women of high social position, not only in the Przeworsk culture, but also in other cultures of Magna Germania. With padlocks it is not so obvious and their function or symbolic meaning could be different. They were found in burials of men (Żerniki / Gross Sürding, grave 9; Karczyn, grave 5), women (Żerniki / Gross Sürding, grave 47; Drăgăneşti Olt, grave 9) and sometimes children (Gavrilovka, grave 31). It is possible that they could be deposited in grave together with closed piece of furniture (a cast), but most probably they have a symbolic meaning. Its function of closing 16 is obvious and it may be treated as an apotropaic specimen to close the dead in grave and protect living ones. 17 There are analogies of such symbolic closing from early medieval, and also later, ethnological sources. Very spectacular example is a very rich barrow-burial of khan of Polovcians by the river Chingul, Zaporozh ye oblast (Отрощенко, Рассамакин 1986: 10, fig. 2. 5; 3; Otroščenko, Rassamakin 1991: 269, fig. 2). Padlocks were deposited on four corners of wooden coffin, most probably as a magical protection. There are many padlocks found on Jewish cemeteries from the sixteenth-nineteenth century, interpreted as magical closure (Jażdżewski 1976: 398; Fijałkowski 2003: 364 nn.). In the nineteenth century, in Polesia (today part of Belorussia) padlocks were deposited in children graves to stop the death in family, to protect their siblings (Ciszewski 1911: 24, 25). Most probably, finds from the Late Roman or Early Migration periods could be interpreted in a similar way. Padlocks found in settlements could be treated as a sign of richness and personal property, which needs protection against possible theft. Found in graves could also be interpreted as an indicators of wealth, but they could have also served as a way of magical closing the dead in a grave, as they were used exclusively for closing. Bibliography Abramek B. Dwa domy ze schyłkowej fazy kultury przeworskiej z Walkowa-Kurnicy, st. 1, gm. Osjaków, woj. łódzkie, datowane metodą radiowęglową // Prace i Materiały Muzeum Archeologicznego i Etnograficznego w Łodzi. 42 ( ). Łódź, Anstee J. W. The Examination, Interpretation and Replication of an Iron Lock from Lullingstone Roman Villa, Kent // The Antiquarian Journal Archeologiczne dziedzictwo Prus Wschodnich w archiwum Feliksa Jakobsona / Aestiorum Hereditas. II. Warszawa, Bednarczyk J., Romańska A. Azylanci problem identyfikacji kulturowej grobów szkieletowych na cmentarzysku w Karczynie, stan. 21/22, gm. Inowrocław, Kujawy // The turbulent epoch new materials from the Late Roman Period and the Migration Period. Krasnobród, Birley A. Security: the keys and locks // Vindolanda. Research reports. New Series. II / II. Greenhead, Brandt J. Eine Siedlung des Jahrhunderts bei Brahlstorf, Ldkr. Ludwigslust // Bodendenkmalpflege in Mecklemburg-Vorpommern, Jahrbuch Ciszewski S. Czaszki ludzkie z kłódkami // Lud XVII. Dąbrowska T. Dwie bogato zdobione popielnice z cmentarzyska kultury przeworskiej w Kamieńczyku, woj. ostrołęckie // Nvnc de Svebis dicendvm est. Warszawa, Dąbrowski K. Sprawozdanie z prac wykopaliskowych we wsi Piwonice, pow. Kalisz w 1953 r. // Materiały Starożyt ne I. Dąbrowski K. Osadnictwo z okresów późnolateńskiego i rzymskiego na stanowisku 1 w Piwonicach, pow. Kalisz // Materiały Starożytne IV. 16 It is opposite to key, which has two functions: opening and closing, and that s why two various symbolic meanings. 17 In grave 46 from Żerniki Wielkie/Gross Sürding, distr. Wrocław a padlock was placed on the neck of the body (fig. 6).

81 Padlocks in the Przeworsk and the Chernyakhov Cultures in the Late Roman Period, as an Evidence of Mutual Contacts 81 Dąbrowski K., Kozłowska R. Dwie osady z okresów późnolateńskiego i rzymskiego we wsi Piwonice, pow. Kalisz // Materiały Starożytne V. Déchelette J. La collection Millon. Antiquités préhistoriques et gallo-romaines. Paris, Dobrzańska H. Osada z późnego okresu rzymskiego w Igołomi, woj. krakowskie. Wrocław; Warszawa; Kraków; Gdańsk; Łódź, Dobrzańska H. Ośrodek produkcji ceramiki siwej z okresu rzymskiego w Zofipolu // 150 lat Muzeum Archeologicznego w Krakowie. Kraków, Dobrzańska H. Roman Period grey pottery production near Cracow: geographical, technological and social dimensions // Bonner Beiträge Fijałkowski P. Obrzędy pogrzebowe Żydów polskich w XVI XIX w. w świetle badań archeologicznych // Kobieta Śmierć Mężczyzna, Funeralia Lednickie, Spotkanie 5. Poznań, Frąsiak P., Gwóźdź B., Siciński W. Wyniki ratowniczych badań archeologicznych na terenie odkrywki Szczerców KWB Bełchatów SA na stan. 12 w miejscowości Chabielice, gm. Szczerców, woj. Łódzkie // Badania archeologiczne na terenie odkrywki Szczerców Kopalni Węgla Brunatnego Bełchatów SA. 1. Łódź, Gáspár D., Salamon A. Rómaikori lakatok magyar múzeumokban // Archaeologiai Értesitő Gindele R. Auf der Töpferscheibe gefertigte römerzeitliche Keramik im oberen Theiss-Becken // Bonner Beiträge Godłowski K. Północni barbarzyńcy i wojny markomańskie w świetle archeologii // Scripta Archaeologica. II. Kraków, Godłowski K. Przemiany kulturowe i osadnicze w południowej i środkowej Polsce w młodszym okresie przedrzymskim i w okresie rzymskim. Wrocław; Warszawa; Kraków; Gdańsk; Łódź, Godłowski K. Jakuszowice. Woiwodschaft Kielce, Gemeinde Kazimierza Wielka, Fundstelle 2. Siedlung der Trzciniec- und Przeworsk-Kultur und des Mittelalters // Recherches archéologiques de Kraków, Grygiel M., Pikulski J., Trojan M. Rescue excavations on the Late Roman period settlement on site 3 in Zagórzyce, com. and distr. Kazimierza Wielka, voiv. świętokrzyskie // Recherches Archeologique. Nov. ser. 1. Kraków, Guillaumet J. P., Laude G. L art de la serrurerie gallo-romaine. L exemple de l agglomération de Vertault (France, Côte-d Or). Dijon, Haberstroh J. Germanische Funde der Kaiser- und Völkerwanderungszeit aus Oberfranken // Materialhefte zur Bayerischen Vorgeschichte. 82. München, Hollack E., Peiser F. E. Das Gräberfeld von Moythienen. Königsberg, Istvánovits E., Kulcsár V. Iranian-Germanic contacts in the Sarmatian barbaricum of the Carpathian basin // Die spätrömische Kaiserzeit und die frühe Völkerwanderungszeit in Mittel- und Osteuropa. Łódź, Jadczyk I. Wyniki badań w Przywozie, pow. Wieluń, na stanowisku 1 i 1a z okresu wpływów rzymskich // Sprawozdania Archeologiczne XXIII. Jażdżewski K. Einiges über kaiserzeitliche, völkerwanderungszeitliche und mittelalterliche Vorhängeschlösser aus Polen // Festschrift Richard Pittioni. II / AAustr. 14. Wien, Kessler P. T. Neue römische Funde aus dem Steinbruch der Portland-Cementfabrik-Heidelberg bei Mainz- Weisenau // Mainzer Zeitschrift. 27. Mainz, Künzl E. Die Alamannenbeute aus dem Rhein bei Neupotz. Plünderungsgut aus dem römischen Gallien. Mainz, Łaszczewska T. Polska środkowa w okresie wędrówek ludów i wczesnego średniowiecza // Prace i Materiały Muzeum Archeologicznego i Etnograficznego w Łodzi. 22. Łódź, Madyda-Legutko R. Die Gürtelschnallen der römischen Kaiserzeit und der frühen Völkerwanderungszeit im mitteleuropäischen Barbaricum / BAR IS Oxford, Makiewicz T. Dalsze badania w Przywozie, pow. Wieluń // Sprawozdania Archeologiczne XXI. Manning W. H. Catalogue of the Romano-British Iron Tools, Fittings and Weapons in the British Museum. London, Nothdurfter J. Die Eisenfunde von Sanzeno im Nonsberg // Römisch-Germanische Forschungen. 38. Mainz, 1979.

82 82 Katarzyna CZARNECKA Otroščenko V. V., Rassamakin Ju. Ja. Der Polovzer-Khan aus dem Čingul -Kurgan // Gold der Steppe. Archäologie der Ukraine. Schleswig, Palade V. Aşezarea şi necropola de la Bârlad Valea Seacâ (sfârşitul sec. al III-lea a doua jumâtate a sec. al V-lea). Bucureşti, Paulík J., Tomčiková K. Ďalši hromadný nález železných predmetov v Plaveckom Podhradí // Slovenska Archeologia LIII / 1. Rodzińska-Nowak J. Jakuszowice stanowisko 2. Ceramika z osady kultury przeworskiej z młodszego i późnego okresu wpływów rzymskich i wczesnej fazy okresu wędrówek ludów. Kraków, Roth H., Theune C. Das frühmittelalterliche Gräberfeld bei Weingarten (Kr. Ravensburg) // Forschungen und Berichte zur Vor und Frühgeschichte in Baden-Würtenberg. 44. Stuttgart, Schmidt B. Die späte Völkerswanderungszeit in Mitteldeutschland. Katalog (Nord- und Ostteil). Berlin, Stadie K. Gräberfeld bei Gr. Sausgarten // Prussia. B. 23 / II. Königsberg, Trohani G., Zorzoliu T. O necropolă din secul IV-lea e. n. descoperită la Drăgăneşti-Olt / Cercetari Arheologice. VI. Bucureşti, Vornic V. Aşezarea şi necropola de tip Sântana de Mureş-Černjachov de la Budeşti. Chişinău, Zoltai L. Die Hügelgräber der römischen Kaiserzeit in Hortobágy // Dissertationes Pannonicae. 2 / 11. Budapest, Zotz L. Die spätgermanische Kultur Schlesiens im Gräberfeld von Gross Sürding. Leipzig, Бapaн B. Д. Памятники черняховской культуры бассейна западного Буга // Древности эпохи сложения восточного славянства / МИА М., Башкатов Ю. Ю. Поселение позднеримского времени Барбара I на юге Среднего Поднепровья // The Turbulent Epoch. New materials from the Late Roman Period and the Migration Period. II / Monumenta Studia Gothica. V. Lublin, Винокур I. С. Історія та культура черняхівських племен Дністро-Дніпровського межиріччя II V ст. н. e. Kиїв, Магомедов Б. Каборга IV (раскопки гг.) // Могильники черняховской культуры. M., Магомедов Б. Черняховская культура. Проблема этносa / Monumenta Studia Gothica. I. Lublin, Отрощенко В. В., Рассамакин Ю. Я. Половецький комплекс чингульского кургану // Археологiя Сымонович Э. А. Раскопки могильника у овчарни совхоза «Приднепровский» на Нижнем Днепре / МИА. 82. М., Храпунов И. Н. Могильник Дружное (III IV вв нашей эры) / Monumenta Studia Gothica. II. Lublin, 2002.

83 Padlocks in the Przeworsk and the Chernyakhov Cultures in the Late Roman Period, as an Evidence of Mutual Contacts 83 Катажина ЧАРНЕЦКАЯ Навесные замки позднеримского времени в пшеворской и черняховской культурах как свидетельство взаимных контактов Резюме Навесные замки, служившие запором для кандалов, вероятнее всего, изобрели кельты. Древнейшие из известных находок относятся к середине Латенского времени (рис. 1. 1). Их заимствовали римляне (рис ), которые вдобавок создали собственные типы этих изделий (рис ). И те, и другие использовались до последних дней Империи. Хотя железные навесные замки, найденные за пределами Лимеса, считают римским импортом, на самом деле почти все они были местного производства. Техническое решение заимствовали у римских вещей, но форма и конструктивные детали были другими (рис. 2). В пшеворской культуре навесные замки известны главным образом на поселениях, в могильниках же найдено всего несколько штук (рис. 3). Они датируются позднеримским временем и началом эпохи Великого переселения народов, фазами C2 D. Навесные замки очень похожей формы и размеров известны на поселениях и могильниках черняховской культуры (культура Синтана-де-Муреш) (рис ), отдельные находки происходят из сарматских могильников, один замок найден в культуре Долькайм-Коврово (самбийско-натангийской) (рис. 5). В вельбарской культуре и культурах эльбского круга навесные замки не известны. Импортные римские навесные замки открыты на поселении Бральсторф (район Лудвигслуст) и на могильнике богачевской культуры Мойтыны во Мронговском повяте. Возможно, навесные замки пшеворской культуры заимствованы из римских провинций вместе с производством кружальной керамики из региона к северу от дунайских провинций, быть может, через группу пшеворских памятников в верхнем Потисье. Схожим образом могли быть также заимствованы навесные замки черняховской культуры. Другая, не подтверждённая, возможность заимствование из греческих городов Причерноморья. Найденные на поселениях навесные замки можно интерпретировать как знак собственности средство защиты от воров. Найденные в могилах вещи, которые могли быть показателями достатка, возможно, имели и другое, символическое значение амулета-апотропея, закрывавшего мертвеца в могиле (рис. 6).

84 84 Charlotte FABECH, Ulf NÄSMAN A Lonely Rider? The Finding Place of the Sösdala Find and the Context of its Finds 1 In the first publication of the Inter Ambo Maria conferences, Maxim Levada discussed the Sösdala find and related European finds in an interesting paper (Levada 2011). We are happy that we got the opportunity to continue the discussion at the second Inter Ambo Maria conference and contribute to the proceedings. We will start with a presentation of the finding place at Sösdala and its local and regional contexts in Scania. The name of Sösdala is more famous then the objects found. That is because Sösdala became the eponym of the Sösdala style when the magnificent bridle was described by John-Elof Forssander in a celebrated paper (Forssander 1937). The Sösdala style is difficult to grasp, but nevertheless one with European significance (e. g. Roth 1979: 55 58; Bitner-Wróblewska 2001: ). Unfortunately, the objects have never been properly published. 2 They and other relevant Scandinavian finds are too often forgotten when the history of the Migration period is written. Consequently, a plausible Scandinavian participation in events during the Migration period is often ignored. We will try to give the Sösdala find an interpretation based on the occurrence of similar finds in Europe and attempt to place the find in an historical context. Does the Sösdala find represent a lonely rider? Find history and find context In June 1929 a man entered the book shop in Sösdala, a small community in central Scania (fig. 1). He asked the bookseller for the value of some metal objects found in the neighbourhood when a group of workers dug in a gravel ridge (fig. 2). Then and there began the rescue of one of the most remarkable Migration period finds from Scandinavia. The bookseller realised the importance of the finds and contacted a local teacher, Mr. Carl Mellton, who had an archaeological interest. Mellton collected the finds from the workers and they told him that the finds came from two depositions. He had the gravel screened to find more objects (fig. 3). When investigating the finding place itself, he found a third deposition with some objects recorded in situ. A lance head of iron was found among stones that the workers had removed from the site. Mellton took some finds to the museum in Lund where they were received by Professor Olof Rydbeck. Olle Källström was sent to the site to make supplementary investigations and take care of the rest of the finds. His report was based on what Mellton and the workers told him. The objects were found a spade spit deep under the surface in three shallow pits. In pit I many decorated mounts of a head bridle of gilded silver and an iron bit were found. Pits II and III contained a large number of bronze mounts from saddles, some of them with silver inlays. No finds were made when Forssander himself visited the site. Gravel digging continued however, and in April 1930 the workers found a fourth pit that contained more saddle mounts of bronze with silver inlays. After 1930 nothing was reported from the site until 1961 when a new find was made ca 60 m to the East, consisting of silver-plated bronze mounts of a second bridle. An archaeological investigation of the site was made and some more objects were 1 Our research is partly funded by the Berit Wallenberg foundation. 2 A full publication is in preparation.

85 A Lonely Rider? The Finding Place of the Sösdala Find and the Context of its Finds 85 Fig. 1. Scania with Sösdala marked. Dotted areas represent the general settled areas in the late first millennium AD Fig. 2. The find spot on the Sösdala gravel ridge. An arrow points to the gravel pit where the finds were made (photo: Carl Mellton, 1929)

86 86 Charlotte FABECH, Ulf NÄSMAN Fig. 3. The Sösdala finds were made in a gravel pit. In order to find more objects the gravel was screened (photo: Carl Mellton, 1929) found (Bunte 1961). The owner of the land now told Carin Bunte that the four pits had been covered by a 5 x 2.5 m large stone layer (possibly the removed stones among which the lance head was found). Interpretation of the depositions For many years the Sösdala depositions were treated as belonging to a group of finds of weapons and/or horse equipment, often deposited in wet areas and labelled weapon sacrifices (Stenberger 1977: ). However, on closer inspection it became obvious that this cannot be the case; the Sösdala finds represent something else. This is also the case with a similar find made at Fulltofta, only 15 km from Sösdala (Fabech 1991a; 1991b). The main argument is that both finds were made on the top of gravel ridges, not in wetlands. And they are not graves; the finds were made close to ground level, not dug down, and human or animal bones were not found (but preservation conditions for organic material are not good). Nothing indicates the presence of graves. Instead, it has been suggested that they represent Totenopfer (funeral sacrifices) or ceremonially deposited horse equipment. The rituals were possibly influenced by nomadic customs originating in Eastern Europe (Fabech 1991a; 1991b). The practice of funeral sacrifices is typical of mounted nomads as Sarmatians and Huns. Such rituals are well known from descriptions of mounted nomads in later periods (Bóna 1979; Tomka 1987). Contacts to peoples in Eastern Europe brought Scandinavians into contact with nomadic cultures in which the horse was central to power and prestige as well as in ritual (Fabech 1991a; 1991b). Perhaps Scandinavian warriors have attended funerary offerings according to nomadic customs or at least heard stories about such rituals. After the burial of a Hun leader, a funeral feast was celebrated in the neighbourhood of the grave. During the ritual weaponry and horse trappings were destroyed, and at last the remains were deposited and covered up. The Fulltofta and Sösdala deposits could represent similar rituals performed on the top of a ridge overlooking the surroundings. At the nearby Vätteryd cemetery, a so-called hammer-shaped bridge-mount of a sword scabbard was found in a destroyed cremation. It is contemporary to the Sösdala bridles and saddles (Strömberg 1961: 88, Taf ; Bemmann 1998: , Fundliste 9; 2006: Fundliste and map,

87 A Lonely Rider? The Finding Place of the Sösdala Find and the Context of its Finds Abb. 6). Maybe the horse equipment of this warrior is found in the depositions on the ridge above. Perhaps the lance head that was found shows that a lance was thrust into the ground to mark the site. The Sösdala finds in its local context In a 10 x 30 km large area around Sösdala, a number of remarkable Migration period finds have been made in a forested area, peripheral to the densely settled plains in south Scania (Carlie 1994). The sites are Claestorp, Fulltofta, Göingeholm, Häglinge, Sjörup and Tormestorp (fig. 4). At Claestorp, ca. 20 km from the Sösdala find spot, another remarkable find was made 1876 in a field (Holmqvist 1961; Arrhenius 1987: ). It is a ca. 4 cm large gold plaque shaped as an eagle with six gold chains attached (fig. 5). Perhaps it once served as a chain connector in a harness or body chain (see Freiberger, Gschwantler 1999: 106, Abb. 23). It is plausibly a late Roman or 87 Byzantine work. It can only be tentatively dated to the Migration period. In a gravel ridge at Fulltofta, ca 15 km from Sösdala, 23 bronze mounts were found 1896 that represent the remains of at least one bridle and one saddle (fig. 6. a b). The mounts resemble the Sösdala saddle mounts and the mounts of the Sösdala 1961 bridle (Fabech 1991a: ). Thus a date in the first half of the fifth century is probable, i. e. Scandinavian period D1, ca AD (Kristoffersen, Magnus 2010: 75 77). During roadwork in 1935 near Göingeholm, ca. 4 km from Sösdala, two clay pots were found four meters apart at the top of a rock outcrop (Arne 1937). A simple pot was empty; the other had ribbed decoration and contained a hoard of jewellery: four silver brooches, one in Nydam style (Bemmann 1998 Fundliste 10), fragments of nine silver clasps, and forty-seven amber beads (fig. 7). The amber beads are probably imports from the south-eastern Fig. 4. Around Sösdala, a number of remarkable Migration period finds have been made. The sites are a cemetery at Vätteryd, a stray find at Claestorp and ritual depositions at Göingeholm, Sjörup, and Tormestorp, and funeral sacrifices at Fulltofta and Sösdala

88 88 Charlotte FABECH, Ulf NÄSMAN Fig. 5. The ca. 4 cm large gold plaque shaped as an eagle is a stray find made 1876 at Claestorp, ca. 20 km from Sösdala (Swedish History Museum, Stockholm: inv. no. 7004; photo: Ulf Bruxe) littoral of the Baltic Sea. Jewellery and pots date the hoard to Scandinavian period D1. The finds do not represent graves, and it is tempting to explain the relation between rock and pots as ritual. Cemeteries near Häglinge, ca. 3 km from Sösdala, were in use from the Migration period to the Viking Age. Near the village a hill called Lunden (= the groove) is covered by stone settings, clearance cairns, and cup-marked stones. At the top a conspicuous formation of basalt is found. Perhaps this was a holy groove. In the High Middle Ages both a parish church and a manor with remains of a stone palace were built at Häglinge. Three times between 1842 and 1859 about 40 fragments of sword mounts and mounts of horse harnesses were found at Sjörup on the shore of Lake Bosarp, ca. 6 km from Sösdala and ca. 2 km from Häglinge (Fabech 1990: ; fig. 2 8; Arrhenius 1987: , Taf. 76). The mounts are cast in silver, gilded, and have chip-carved ornamentation (fig. 8). Many are nielloed and some have set stones. A few mounts are decorated in Scandinavian style I but most mounts seem to be decorated in the earlier Nydam-style. The last objects in the deposition can be dated to the third quarter of the fifth century AD or Scandinavian period D2a. Some of the buckles resemble so-called Gothic buckles. Too little is known about the find context to interpret the find as either a treasured hoard, trophies hidden after battle or a votive offering. At Tormestorp, ca. 12 km north of Sösdala, a gilded silver buckle was found at the shore of Lake Finja in It is decorated with geometrical chipcarving in Nydam style. Also a silver-plated hammer-shaped bridge mount of bronze from a sword scabbard was found (fig. 9) (Strömberg 1961: 73, Taf a b; Bemmann 1998: Fundlisten 9, 10).

89 A Lonely Rider? The Finding Place of the Sösdala Find and the Context of its Finds 89 a Fig. 6. Examples of bridle mounts deposited at Fulltofta, ca. 15 km from Sösdala. a a two-way strap junction of bronze with silver details, decorated with a pelta-shaped silver pendant, b two four-way strap junctions of bronze with silver details (Lund University Historical Museum: inv. no. J 14080; photo: Charlotte Fabech) b

90 90 Charlotte FABECH, Ulf NÄSMAN Fig. 7. A clay pot with ribbed decoration was found in 1935 near Göingeholm, ca. 4 km from Sösdala. It contained a hoard of jewellery: four silver brooches, one in Nydam style, fragments of nine silver clasps, and forty-seven amber beads (1936 photo, Swedish History Museum: inv. no )

91 A Lonely Rider? The Finding Place of the Sösdala Find and the Context of its Finds 91 Fig. 8. Between 1842 and 1859 sword mounts and mounts of horse harnesses were found at Sjörup, ca. 6 km from Sösdala. The mounts are cast in silver, gilded, and have chip-carved ornamentation in Scandinavian Nydam-style and animal style I (Swedish History Museum: inv. no. 2663; photo: Sören Hallgren)

92 92 Charlotte FABECH, Ulf NÄSMAN a b Fig. 9. A gilded silver buckle and a silver-plated hammer-shaped bridge mount of bronze from a sword scabbard found at Tormestorp, ca 12 km north of Sösdala (Lund University Historical Museum: no inv. no.; a Strömberg 1951: Taf a; b photo: Charlotte Fabech) No more finds were made during a metal detecting survey the same year. Both objects can be dated to Scandinavian period D1. Many of the artefacts in these finds symbolise high social standing. The find contexts reflect ritual behaviour. Characteristic of the social landscape in this period is the interconnectedness of power and religion. The finds in the landscape between the find-spots Claestorp, Tormestorp, and Sjörup indicate that an elite had established itself here in the middle of Scania already in the early Migration period. A possible continuity into the High Middle Ages is suggested by both a manor at Häglinge with remains of a stone palace and a royal domain Mjölkalånga-Sösdala, donated to Roskilde cathedral, Zealand, in the eleventh century. Finds from cemeteries in the area demonstrate that people lived here continuously. But knowledge about settlements is lacking from the Migration period as well as later. These finds in a forested area in the middle of Scania demonstrate considerable mobility of people. Those who lived here had contacts to the European martial elite, in which several gentes were included a Gothic multicultural and polyglot setting with strains of nomadic cultures (Bemmann 2007). This group of finds dates to a period of fun-

93 A Lonely Rider? The Finding Place of the Sösdala Find and the Context of its Finds damental changes in Europe, (see below). The Sösdala finds The finds made at Sösdala now consist of about 280 metal parts from five pits along the top of a gravel ridge. We know that the five depositions that 93 were recorded contained remains of two bridles and five saddles. This is a minimum number, since we do not know whether finds were lost before Nor do we know whether further depositions were found but not reported between 1930 and What has happened at the site after 1961? The horse bits are made of iron, but all mounts are made of either bronze or silver. The saddles and bridles were violently destroyed and cut to pieces before deposition. When found bits of leather were still attached to some mounts. The objects were not burnt before deposition. Fig. 10. Sösdala , strong bronze rings belong to the saddle remains, one with inlays of silver wire (all Sösdala finds are kept in Lund University Historical Museum, inv. no. J 25570; photos: Charlotte Fabech) The saddles Strong bronze rings as those found at Sösdala are characteristic of saddle remains found in Denmark and Sweden (Ramqvist 1992: 84, 86). It is assumed that the rings were used to fasten equipment to the saddles (fig. 10). Another characteristic trait of Scandinavian ring saddles is that strong mounts held two bows in place, one in front and one at the back of the saddle (fig. 11). Such strong bronze mounts are at present only found in Denmark and Sweden. Many of the Sösdala rings and mounts have inlays of silver wire (fig.12). Some mounts are partially covered by silver. Bronze nails, some with a silver cap, fastened the mounts to the wooden frame. This type of parade saddle seems to be Fig. 11. Rune Norberg s reconstruction of a saddle of Scandinavian type. Note the saddle rings and the mounts that hold the two bows in place (Norberg 1931: Abb. 14)

94 94 Charlotte FABECH, Ulf NÄSMAN of South Scandinavian origin. But it is generally assumed that the basic wooden construction can be derived from saddles found in the nomadic contexts of Eastern Europe. Certainly, the Scandinavian ring saddles seem to have had the same wooden construction as Migration period saddles known from several finds that usually are attributed to the Huns. But so-called Hun saddles lack the mounts and rings characteristic of Scandinavian saddles (Anke 1998: 120; 2007: 224). And in their turn, the Scandinavian saddles are never decorated with gold foils as the Hun saddles are. In fact, saddles with bow-mounts and rings are known in South Scandinavia long before the Huns invaded Europe, and can be followed back as far as to the early third century (Ørsnes 1993: ; Carnap-Bornheim, Ilkjær 1996: ). However, the early ring saddles lack the strong bronze mounts of the front and rear bows that characterize the Migration period saddles. But they reveal that the saddle mounts of Sösdala represent a Scandinavian development of a probably nomadic saddle type that came to Scandinavia already in the third century. Only in two finds do we have information about find contexts that tell us about chronology: Högom and Sösdala. Only the grave at Högom is a truly closed context (Ramqvist 1992). It is dated by Ramqvist to around AD 500 (Ramqvist 1992: 221), but he does not phase it in relation to a relative-chronological system. The weapons in Högom can be placed early in Scandinavian period D2 (Fabech 1987: 43) what in Rau s recent chronological analysis correspond to Scandinavian period D2a, ca. 455/ /515 AD. The five Sösdala depositions of mounts from two bridles and five saddles indicate that similar saddle mounts were in use already during Scandinavian period D1, ca /465. Fig. 12. Sösdala ; one of the saddle bow mounts. Note inlays of silver wire, animal heads covered by silver, and bronze nails with silver caps The parade bridle 1929 In the material from there seems to be remains of only one bridle, but it is truly a parade bridle. Most of the objects found in pit I belong to this bridle, but possibly also mounts of horse body harness are represented. The bridle was brutally cut into pieces during the rituals that preceded the deposition. But most of it seems to be preserved. The bit is made of iron but all mounts of silver (fig. 13). Several are gilded and stamp-decorated. Niello is

95 A Lonely Rider? The Finding Place of the Sösdala Find and the Context of its Finds Fig. 13. Sösdala 1929, the bi-metallic bit of iron and silver 95 used very restrictedly and only found on the brow mount pendant. The iron bit is fixed to the bridle rings by heavy mounts. Reins and straps are fastened to the bridle rings by four beautifully crafted mounts, two simpler and two heavier, which end in naturalistically shaped, double horse-heads in profile (fig. 14). The straps of the head bridle were decorated with four open-worked pendants of very high quality (fig. 15). The beautiful brow mount with double horse heads in profile has an openworked pendant with a nielloed pattern of trefoil leaves (fig. 16). The straps have been adorned with about forty gilded mounts; some of them were united by ring-link connectors (fig. 17). Two different shapes of strap-ends are found, two of each. Three buckles are identical but the frame of two of them has been deformed (fig. 18). A fourth, differently shaped buckle may have completed the set. Possibly a mount decorated with a Sösdala star with a pelta-shaped pendant (fig. 19), also belonged to the head bridle but perhaps it adorned a breast girth. A fifth buckle for a broader strap has no tongue and its function is at present unclear to us (fig. 18), as is the function of a number of simpler mounts. Most of the gilded silver mounts of the head bridle from 1929 are of a very high quality. But not all mounts were made by the same craftsmen or in the same workshop. The bridle has been used and it is worn. Broken or missing pieces have been replaced by simpler copies (fig. 15). The stamps used to decorate the Sösdala mounts are very varied and some are quite unusual on Scandinavian objects decorated in the so-called Sösdala style (fig. 17). For the date, see below. Fig. 14. Sösdala 1929, one of two mounts that fastened the head bridle to the bridle rings The bridle 1961 Remains of a second bridle were found in 1961 (Bunte 1961). The bridle bit is of iron (fig. 20). This bridle is much simpler than the bridle The excavator attempted a reconstruction of the head bridle (fig. 21). All mounts are made of bronze. Many are partially covered by silver and silver wire is inlayed in some mounts. Technically the mounts of this bridle resembles some of the saddle mounts. The straps of the head-bridle seem to have been divided in shorter strips that were linked to one another by ring-link connectors. One of these carries a simple pelta-shaped pendent. Two buckles are part of the bridle. A narrow pendant ends in a stylised animal head and

96 96 Charlotte FABECH, Ulf NÄSMAN a b Fig. 15. Sösdala 1929, two of four open-worked pendants of the head bridle: a is an original of high quality, b is a copy of lower quality possibly served as strap-end. The Sösdala 1961 bridle has great similarities to the bridle mounts in the Fulltofta find (Norberg 1931). The decoration of the pelta-shaped pedant dates the find to Scandinavian period D1, a dating probably also relevant for the Sösdala 1961 bridle. The provenience The saddle mounts seem all to be an output of South Scandinavian workshops. Nothing indicates that they were made outside southern Scandinavia. But the mounts of the two bridles 1929 and 1961 belong to functional types with broader European distributions. Similar bridle

97 Fig. 16. Sösdala 1929, the brow mount is a three-way connector with an attached open-worked pendant with a nielloed pattern Fig. 17. Sösdala 1929, three of the gilded strap mounts and one of the strap-ends

98 98 Charlotte FABECH, Ulf NÄSMAN Fig. 18. Sösdala 1929, two of the strap buckles Fig. 19. Sösdala 1929, a mount decorated with pelta-shaped pendant Fig. 20. Sösdala 1961, bi-metallic bridle bit of bronze and iron with rings and some of the mounts of the head bridle. Some parts are covered by silver (photo: Charlotte Fabech)

99 A Lonely Rider? The Finding Place of the Sösdala Find and the Context of its Finds 99 mounts dated to Scandinavian period D1 are listed by Ørsnes (Ørsnes 1993: ). The finds are in Denmark: Dallerup Sø (we should add Hessel and Hellum Bakker), and in Sweden: Finnestorp, Fulltofta, Sösdala 1929, Sösdala 1961, Vennebo, and Åmossarna. According to their appearance most of these finds could represent the output of Scandinavian workshops. But the bridle Sösdala 1929 does not fit in easily. How many similar bridles do we know? Where was this bridle produced? In Scania or elsewhere? By native or foreign craftsmen? Forssander suggested that the bridle mounts are Erzeugnisse einer nordischen, warum nicht schonischen Metallindustrie [the product of a Nordic, why not Scanian metal industry]. He continued anyhow to explore influences from what he called Gothic-Hungarian elements and links to South-eastern Europe (Forssander 1937: ). This idea was rejected by Helmut Roth, who argued that the Sösdala style was created in the interaction zone along the River Rhine between Roman and Germanic. According to him the Sösdala style came to Scandinavia over the Saxon area. Economic, social and political differences in western, central, southern, and northern Europe during the period when the bulk of the archaeological record was formed, explain some of the difficulties we have today when we try to interpret variations in the archaeological record of different European regions. But nevertheless, the known distributions seem to be reliable and give an impression that the original distribution of objects decorated in Sösdala style is mainly West European while the distribution of bridle mounts of the Sösdala Untersiebenbrunn Coşoveni group is mainly east European (see below). The star-decorated pelta-shaped pendant is certainly made in a Scandinavian workshop, but Roth is not certain that all the other Sösdala mounts are Germanic (Roth 1979: 55 58; also Näsman 1984a: 104). We agree that the uniqueness of some mounts of the bridle Sösdala 1929 compared to objects in typical Scandinavian Sösdala style makes it worthwhile to discuss the possibility of a provenience outside of Scania. Fig. 21. Sösdala 1961, a reconstruction of the head bridle (Bunte 1961: fig. 10) The Sösdala-Untersiebenbrunn-Coşoveni group The Sösdala style is difficult to grasp. As Anna Bitner-Wróblewska expressed it on the one hand there is a horizon of specific artefacts, on the other a largely undefined ornamental style (1991: 89). Instead Anna Bitner-Wróblewska suggests that we use the concept Sösdala horizon in our chronological discussions. The distribution of objects related to the Sösdala bridle 1929 make it relevant to extend the Sösdala concept outside Scandinavia. Helmut Roth labelled the phenomenon the Sösdala- Coşoveni style (Roth 1979: 55 58) and another label is Sösdala-Untersiebenbrunn style (Nothnagel 2008: 40). We will not enter a discussion about style here. To discuss where, when, and how the Sösdala style developed is not central for understanding the provenience of the Sösdala 1929 bridle, and this was already noted by Forssander. So perhaps it is best to exclude style from the concept and instead we will focus on bridle mounts and use the geographical definition the Sösdala Untersiebenbrunn Coşoveni group (Nothnagel 2008: 166). Many different techniques were used to shape and decorate the Sösdala bridle mounts: silver was cast, silver-plate was cut and shaped, decoration was stamped, chip carved, gilded, nielloed, etc. Few Scandinavian finds are of the same out-standing quality as the Sösdala bridle. As far as we are aware, only a mount with double horse heads on a

100 100 Charlotte FABECH, Ulf NÄSMAN sword scabbard from Øvre Veien in Norway have a similar quality (Arwidsson 1984). According to Jan Bemmann the decoration of the Veien mounts belong to the Nydam style (Bemmann 1998: Fundliste ), but in our opinion there is also a strong link to the Sösdala mounts. Be that as it may, the Sösdala and Nydam styles are more or less contemporary and belong to Scandinavian period D1. A dating of the Sösdala bridle 1929 to the first half of the fifth century is based on the dating of the Sösdala style (Näsman 1984b: 60 71; cf. Bemmann 1998: ). In a recent Scandinavian Migration period chronology, this corresponds to Scandinavian period D1, ca AD (Kristoffersen, Magnus 2010: 75 77). In the opinion of Andreas Rau some graves of Bitner-Wróblewska s Sösdala horizon that belong to phase C3 of the Late Roman Iron Age should be excluded from the Sösdala-Untersiebenbrunn group (Rau 2010: 63). Related continental finds as Untersiebenbrunn; Kachin, and Coşoveni belong to Central European period D2, in Eastern Europe including the so-called Untersiebenbrunn/ Kachin Sinyavak horizon. Continental D2 is dated to the period 380/ /450 (Tejral 1997: 351; Nothnagel 2008: 37 39). Recently Scandinavian period D1 is dated to ca /465 (Rau 2010: ). At present we see no possibilities to get any closer about the date of the making of the Sösdala 1929 bridle. New excavations at Finnestorp have produced new evidence of equestrian equipment contemporary to the Sösdala finds (Nordqvist 2005). Some mounts have similarities to the Sösdala mounts, but the find context is very different. Finnestorp is an offering in wetland of military gear. Some weapons have been found including sword mounts. More significant finds in this context are remains of bridles as well as saddles. Mounts decorated in Sösdala and Nydam style as well as Style I represent horse trappings as well as sword mounts that is both the warriors and their horses. Ritual depositions at Finnestorp were made one or more times in the fifth century; find types date to Scandinavian period D1 and D2a. The European connection The best parallel to the Sösdala 1929 bridle is not found in Scandinavia, but in Ukraine at Kachin (Kazanski 1992: fig. 9; Levada 2011: fig. 3). The find consists of both bridle mounts and silver sheet brooches. Its context is believed to be a hoard. Both Sösdala and Kachin belong to a group of exclusive finds, which represent the superregional mounted warrior elite that developed in the early Migration period. Characteristics are bimetallic bridle bits, mounts of bronze or silver, which are partially gilded or covered by silver, and stamped ornamentation in Sösdala Untersiebenbrunn sty le. According to Jaroslav Tejral the finds from Unter siebenbrunn and Kachin belong to a phase in which east European influences merged with Central European late Antique culture and a horizon of widely distributed relatively homogenous material appeared (Tejral 1988: 241). The bridles are constituents of the material culture of peoples living in Central and Eastern Europe, outside the disintegrating Roman Empire. The finds from Untersiebenbrunn, Jakuszowice, Kachin, Bar, and Coşoveni come from graves or hoards (Levada 2011). The Scandinavian finds from Finnestorp, Fulltofta, Sösdala, and Vennebo are all ritual depositions, not graves, nor hoards. In spite of great similarities, the obvious variations in shape and decoration of the bridle mounts indicate that we only know a small part of what once existed. This is reflected by South Scandinavian finds of incomplete bridles (Bregentved, Dallerup, Hellum bakke, Hessel and Åmossarna, see Ørsnes 1993). Four finds of axe-shaped pendants similar to pendants in Untersiebenbrunn and Coşoveni underline this (two come from Roman forts at Gigen in Bulgaria and Grundremmingen in Bavaria, a stray find is from Dahmker, Lauenburg, North Germany, and one is from an unknown site in Hungary (Nothnagel 2008: 171, Taf. 28). It is consequently difficult to find out where in the vast territory from Austria and Ukraine in the South to Sweden in the North that the different bridles were made, but a number of workshops must have been active (fig. 22). Late Roman arts and crafts made a strong impact on the fashion of the multicultural martial elite outside the Empire and this is visible in weaponry as well as in horse equipment. It has been suggested that bridles like those from Untersiebenbrunn and Coşoveni were produced in provincial-roman workshops working for the eastern gentes, which were allied with the Roman Empire (Tejral 2001: 40; Nothnagel 2008: ). The finds from Kachin, Bar, and Jakuszowice demonstrate that the

101 A Lonely Rider? The Finding Place of the Sösdala Find and the Context of its Finds 101 Fig. 22. A map of the distribution of finds related to the Sösdala 1929 bridle Fig. 23. A three-way connector in the silver hoard from Bar in Ukraine can be from a brow mount as that found in Sösdala 1929 (fig. 8) (Levada 2011: fig. 6) style rapidly spread in Eastern Europe, but the bridles were plausibly locally made. The Scandinavian finds reveal that Scandinavian warriors had close contacts to the elite of Central and Eastern Europe. The question is, whether it was a transfer of objects or ideas? Remarkable similarities between the Scandinavian and Continental finds, but also the dissimilarities, are important when discussing where the bridles were made. The three-way connector of the brow mount from Sösdala has a close counterpart in the hoard from Bar in Ukraine (fig. 23). The bird s head found in the Bar hoard is possibly a fragment of a strap junction similar to the mounts from Sjörup, in Scania (fig. 8) (Arrhenius 1987: Taf. 76; Bemmann 2007: 181). It is probably no coincidence that Sjörup is situated only six kilometres from Sösdala. A shield-shaped strap-junction from Finnestorp also reveals the Scandinavians close connections to the Continental elites (Nordqvist 2005: Abb. 7. 5). With its faceted shield boss it is very similar to four round strap-junctions from Untersiebenbrunn in Austria (Stadler 1987: Taf. 50; Kazanski 1992: 81; Nothnagel 2008: Taf ). Faceted shield

102 102 Charlotte FABECH, Ulf NÄSMAN bosses date to the early Migration period and are found in Central and Eastern Europe but not in Scandinavia (Kazanski 1988: 76, fig. 6 7). The shield of Stilicho in the famous Monza diptych (Conti 1999: 14 16) carries a faceted shield boss what underlines both chronology and social setting. Mounts of the bridle Sösdala 1929 may have been made somewhere in South Scandinavia or Central and Eastern Europe, an area roughly circumscribed by a line connecting the find spots Untersiebenbrunn Coşoveni Bar Kachin Sösda la Un tersiebenbrunn. To get any closer will need new and detailed investigations. The find spots Untersiebenbrunn, Coşoveni, Bar, Kachin, and Sösdala are all placed within the distribution of various types of green glass beakers decorated with wheel-cut ovals and / or facets (Näsman 1984: maps 3, 5 6; Straume 1987: Karten 2, 4, 5; Gavritukhin 2011: fig. 1, 6; Magomedov 2011: fig. 1). Dated beakers are mostly from period C2 C3 but in Scandinavia also from Scandinavian period D1 D2; many of the late vessels are repaired and were plausibly old when put in the grave. The most important routes, along which glass vessels were distributed in the fourth century, went probably north of the Carpathians. The routes seem to have lost their importance in the fifth century. But connections from the Black Sea area to central Europe continued along the Danube (Straume 1987: 72; Magomedov 2011: 180). However, the distributions of glass vessels and horse bridles are probably not linked to one another but rather two different mirror images of a third phenomenon. At pre sent it seems hazardous to suggest that the two hoards with horse equipment at Bar and Kachin indicate that the old communication corridor between the Black Sea and the Baltic north of the Carpathians was still open in the fifth century. They rather indicate the eastern part of an interaction zone where Late Roman, Germanic, and Nomadic cultures met in the early Migration period. A sketch of the historical context The material culture of the multicultural warrior elite of the fifth century in Scandinavia, to which the Sösdala finds belong, reveals that the influences from the Roman Empire were filtered through Germanic peoples like Saxons and Goths but also that it was influenced by nomadic cultures like Alans and Huns. The Central European period D1 ca. 360/ /410 AD corresponds to the early Hunnic period when the Hunnic centre was still east of the Carpathians. In Central European period D2 ca. 380/ /450 AD the Hun centre moved into the Danubian basin (Bóna 1991). The horse bridles discussed in this paper belong to this historic period and consequently reflect the socio-political and military processes of this time. The fifth century was in Scandinavia a period during which several wars were fought as demonstrated by a number of offerings of military gear in bogs, distributed from the Danish-German borderland to central Sweden (Fabech 1996: 137; Bemmann, Bemmann 1998: 339, Abb. 144). Many Scandinavian finds from the fifth century evince that Scandinavian warriors also took part in the conflicts on the Continent. Solidi hoards found in eastern Scandinavia are for instance witness of soldiers pay and some date to Scandinavian period D1 (Fischer et al. 2011: 201). Probably Scandinavian warriors formed part of the military units at contemporary battle-fields. Scandinavian mounted warriors of the period rode richly equipped horses and carried swords that could match those of their European counterparts. Through their participation in the Roman army or in one of the barbarian hosts, young members of the Scandinavian elite got other experiences than those who stayed at home. Plausibly groups of experienced warriors eventually returned home and tried to find a place fitting their newly won merits and new ideas, ideas that carried with them the seeds of future conflicts with those who lived according to the traditional values of the tribe. If leaders at home could not accept their new ideas, they can have found their own way with weapons in hand. A lonely rider? The Sösdala 1929 bridle is exceptional but the horseman, whose horse equipment was deposited here, was not a lonely rider! He was but one of several elite warriors who returned home with new ideas. In this case, the leaders in the central settlement areas of southern Scania seem to have been able to divert the force of returning warriors to tasks in a more peripheral part of Scania.

103 A Lonely Rider? The Finding Place of the Sösdala Find and the Context of its Finds 103 Bibliography Anke B. Studien zur reiternomadischen Kultur des 4. bis. 5. Jahrhunderts 1 2. Berlin; Weisbach: Anke B. Der reiternomadische Steppenkriger // Attila und die Hunnen. Speyer; Stuttgart, Arne T. J. Ett skånskt fynd från folkvandringstiden // Från stenålder till rokoko. Studier tillägnade Otto Rydbeck. Lund, Arrhenius B. Skandinavien und Osteuropa in der Völkerwanderungszeit // Germanen, Hunnen und Awaren. Schätze der Völkerwanderungszeit. Nürnberg, Arwidsson G. The splendid sword from Veien in Norderhov // Festskrift til Thorleif Sjøvold / Universitetets Oldsaksamlings skrifter. 1. Oslo, Bemmann G., Bemmann J. Der Opferplatz von Nydam. Die Fund aus den älteren Grabungen: Nydam-I and -II Neumünster: Bemmann J. Der Nydam-II-Fund // Bemmann G., Bemmann J. Der Opferplatz von Nydam. Die Fund aus den älteren Grabungen: Nydam-I and -II Neumünster, Bemmann J. Eine völkerwanderungszeitliche Bestattung aus Epöl, Kom. Esztergom, mit Schwertriemendurchzügen skandinavischer Form // Miscellane Romano-Barbarica in honorem septagenarii magistri Ion Ioniţă oblata. Iaşi, Bemmann J. Hinweise auf Kontakte zwischen dem hunnischen Herrschaftsbereich in Südosteuropa und dem Norden // Attila und die Hunnen. Speyer; Stuttgart, Bitner-Wróblewska A. From Samland to Rogaland. East-west connections in the Baltic basin during the Early Migration period. Warszawa, Bóna I. Die Archäologischen Denkmäler der Hunnen und der Hunnenzeit in Ungarn. Spiegel der internationalen Hunnenforschung // Niebelungenlied / Ausstellungskatalog des Vorarlberger Landesmuseum. 86. Bregenze, Bóna I. Das Hunnenreich. Stuttgart, Bunte C. A New Bridle Find from Sösdala // Meddelanden från Lunds universitets historiska museum Lund, Carlie A. På arkeologins bakgård. En bebyggelsesarkeologisk undersökning i norra Skånes inland baserad på synliga gravar / Acta Archaeologica Lundesia, Ser. In Lund, Carnap-Bornheim C. von, Ilkjær J. Illerup Ådal Die Prachtausrüstungen / Jysk arkæologisk selskabs skrifter. 25 / 5 7. Højbjerg; Århus, Conti R. Il tesoro. Guida alla conoscenza del Tesoro del Duomo di Monza. Monza, Fabech Ch. Sjörup an old problem in a new light // Meddelanden från Lunds Universitets Historiska Museum. New series ( ). Fabech Ch. Neue Perspektiven zu den Funden von Sösdala und Fulltofta // Studien zur Sachsenforschung. 1991a. 7. Fabech Ch. Booty sacrifices in Southern Scandinavia: a reassessment // Sacred and profane / Oxford University. Committee for archaeology. Monograph. 32. Oxford, 1991b. Fabech Ch. Booty sacrifices in Southern Scandinavia a history of warfare and ideology // Roman Reflections in Scandinavia. Rome, Fischer S., López Sánchez F., Victor H. The 5th century hoard of Theodosian solidi from Stora Brunneby, Öland, Sweden // Fornvännen Forssander J.-E. Provinzialrömisches und germanisches. Stilstudien zu den schonischen Funden von Sösdala und Sjörup // Kungl. Humanistiska vetenskapssamfundet i Lund. Årsberättelse / Meddelanden från Lunds universitets historiska museum Freiberger V., Gschwantler K. Beobachtungen zur Herstellungstechnik und Tragwise der Goldkette // Barbarenschmuck und Römergold. Der Schatz von Szilágysomlyó. Wien, 1999.

104 104 Charlotte FABECH, Ulf NÄSMAN Gavritukhin I. Cut Glass Beakers within the Context of Studies on the Connections between the South of Eastern Europe and Scandinavia in the Late Period of Roman Influence and the Great Migration Period // Inter Ambo Maria. Contacts between Scandinavia and the Crimea in the Roman Period. Kristiansand; Simferopol, Holmqvist W. En främmande fågel // Fornvännen Kazanski M. Quelques paralleles entre l armement en Occident et a Byzance (IV e VII e s.) // Gaule merovingienne et monde mediterraneen. Lattes, Kazanski M. A propos des armes er les éléments de harnachement orienteaux en Occident à l époque des Grandes Migrations (I Ve Ve s.) // Journal of Roman Archaeology Kazanski M. Les Goths et les Huns. A propos des relations entre les Barbares sédentaires et les noma des // Archéologie medieval Kristoffersen S., Magnus B. Spannformete kar. Udvikling og variasjion / AmS-Varia. 50. Stavanger, Levada M. To Europe via the Crimea: on possible migration routes of the northern people in the Great Migation period // Inter Ambo Maria. Contacts between Scandinavia and the Crimea in the Roman Period. Kristiansand; Simferopol, Magomedov B. The Chernykhov People s Contacts with Scandinavia and the Crimea // Inter Ambo Maria. Contacts between Scandinavia and the Crimea in the Roman Period. Kristiansand; Simferopol: Näsman U. Glas och handel i senromersk tid och folkvandringstid / Aun. 5. Uppsala, 1984a. Näsman U. Zwei Relieffibeln von der Insel Öland // Praehistorische Zeitschrift. 1984b. 59 / 1. Norberg R. Moor- und Depotfunde aus dem 5. Jahrhundert nach Chr. in Schonen // Acta Archaeolo gica Nordqvist B. Der Krigesbeuteopferplatz von Finnestorp in Schweden // Offa ( ). Nothnagel M. Die völkerwanderungszeitlichen Bestattungen von Untersiebenbrunn, Niederösterreich (Diplomarbeit). Wien, Ørsnes M. Zaumzeugfunde des Jhrh. nach Chr. in Mittel- und Nordeuropa // Acta archaeologica / 2. Ramqvist P. H. Högom. The excavations / Archaeology and environment. 13. Umeå, Rau A. Nydam mose 1-2. Die personengebundenen Gegenstände. Grabungen Åhus; Højbjerg, Roth H. et alii. Kunst der Völkerwanderungszeit / Propyläen Kunstgeschichte. Supplementband 4. Frankfurt am Main; Berlin; Wien, Stadler P. Die Bevölkerungsstrukturen nach Eugippius and den archäologischen Quellen. // Germanen, Hunnen und Awaren. Schätze der Völkerwanderungszeit. Nürnberg, Stenberger M. Nordische Vorzeit 4. Vorgeschichte Schwedens. Neumünster, Straume E. Gläser mit Facettenschliff aus skandinavischen Gräbern des 4. und 5. Jahrhunderts n. Chr. Oslo, Strömberg M. Untersuchungen zur jüngeren Eisenzeit in Schonen 1 2 / Acta Archaeologica Lundensia, ser. in Lund, Tejral J. Zur Chronologie der frühen Völkerwanderungszeit im mittleren Donauraum // Archaeologia Austriaca Tejral J. Neue Aspekte der frühvölkerwanderungszeitlichen Chronologie im Mitteldonauraum // Neue Beiträge zur Erforschung der Spätantike im mittleren Donauraum. Brno, Tejral J. Die östlichen Foederaten in Pannonien // Das Gold der Barbarenfürsten. Stuttgart, Tomka P. Der hunnische Fundkomplex von Pannonhalma // Germanen, Hunnen und Awaren. Schätze der Völkerwanderungszeit. Nürnberg, 1987.

105 A Lonely Rider? The Finding Place of the Sösdala Find and the Context of its Finds 105 Шарлотта ФАБЕК, Ульф НЕСМАН Одинокий всадник? О месте, где была обнаружена находка из Сёсдалы, и о контексте её содержимого В статье рассказывается о месте, где была обнаружена знаменитая находка из Сёсдалы, а также о её местном и региональном контексте в рамках региона Скания. Имя Сёсдалы знаменито благодаря украшениям драгоценной уздечки, обнаруженной в 1929 г. и ставшей эпонимным памятником для одноимённого стиля. К сожалению, находка никогда не публиковалось на должном уровне. В данной статье делается попытка интерпретировать находку из Сёсдалы на основании подобных памятников на территории Европы и поместить её в исторический контекст. Но скрывается ли за находкой из Сёсдалы одинокий всадник? На протяжении многих лет находка из Сёсдалы понималась как одна из группы памятников, находившихся во влажной среде и отмеченных жертвоприношениями оружия. Однако, находка из Сёсдалы и близкая ей из близлежащей Фультофты представляют собой нечто иное. Обе находились на вершине гравийных гряд, а не в заболоченной местности. Ничто не указывает на существование могил. Предполагалось, что они представляют собой Totenopfer (погребальные приношения) или «конскую сбрую, принесённую в жертву в церемониальных целях». На ритуалы, вероятно, оказали влияние кочевые традиции, происходившие из Восточной Европы. Недалеко от Сёсдалы, в обширном лесном массиве 10 х 30 км на периферии плотно населённых равнин южной Скании, сделаны выдающиеся находки Эпохи великого переселения народов в Клесторпе, Фультофте, Гёйнгехольме, Хеглинге, Шёрупе и Торнесторпе. Эти находки, отражающие ритуальные действия, символизируют высокий общественный статус. Множество необычных вещей указывают на то, что здесь, в центре Скани, в начале эпохи Великого переселения утвердилась некая группа знати. На возможную преемственность с эпохой Высокого Средневековья указывает поместье с каменным дворцом и королевским доменом, известное с XI в. Тем не менее, сведения о поселениях отсутствуют. Сегодня сделанные в Сёсдалу находки вклю чают около 280 металлических изделий из пяти ям, располагавшихся на вершине гравийной гряды. В этих пяти местах найдены остатки двух уздечек и пяти сёдел. Удила железные, но все накладки выполнены из бронзы или серебра. Сёдла и уздечки были преднамеренно испорчены и разрублены на куски, но воздействию огня перед закапыванием в землю не подвергались. Сёдла с бронзовыми кольцами и оковками лук, как представляется, имеют скандинавское происхождение. Скандинавские сёдла с кольцевидными накладками имели такую же деревянную основу, как и сёдла кочевых народов Восточной Европы, но у так называемых «гуннских» сёдел отсутствуют накладки и кольца, характерные для скандинавских сёдел. Сёдла с оковками лук и кольцами известны в южной Скандинавии с начала III в. Таким образом, накладки сёдел типа «Сёсдала» представляют собой скандинавское развитие седла, по всей вероятности, кочевого типа, которое попало в Скандинавию к III в. Большинство накладок на оголовье уздечки из позолоченного серебра 1929 г. очень высокого качества. Штампы, использовавшиеся для декорирования многих из этих накладок, очень разнообразны, некоторые из них довольно необычны для скандинавских вещей, декорированных в так называемом «стиле Сёсдала». Вторая уздечка, обнаруженная в 1961 г., гораздо проще, а накладки деланы из бронзы, отдельные их детали из серебра. Уздечка из Сёсдалы 1961 г. имеет аналогии на деталях узды из Фультофты. Накладки на уздечки из Сёсдалы относятся к функциональным типам, широко распространённым в Европе, но уздечка 1929 г. из Сёсдалы не такова. уникальность этой узды заставляет задуматься о её возмож-

106 106 Charlotte FABECH, Ulf NÄSMAN ном происхождении за пределами Скании. Датировка этой узды первой половиной V в. основана на датировке стиля «Сёсдала» периодом D1, то есть приблизительно /465 гг. Близкие находки на Континенте, в Унтерзибенбрунне, Качине и Кошовени относятся к периоду D2 центрально-европейской шкалы, а в Восточной Европе датируются 380/ /450 гг. Наилучшая аналогия узде из Сёсдала происходит не из Скандинавии, а из Качина на Украине. Близкие находки из Унтерзибенбрунна (Австрии), Кошовени (Румыния), Якушовице (Польша), а также Бара и Качина (Украина) указывают на восточную часть зоны, где в начал эпохи Великого переселения позднеримская культура взаимодействовала с германскими и кочевыми культурами. Находка из Сёсдала показывает, что скандинавские воины имели тесные связи с центрально- и восточноевропейской знатью. Центр державы гуннов переместился в Подунавье и Центральную Европу в период D2, то есть около 380/ /450 гг. Уздечка коня из Сёсдалы, найденная в 1929 г., относится к этому историческому времени и связана с происходившими тогда военными процессами. Археологические находки демонстрируют, что скандинавские воины, вероятно, составляли часть военных формирований, участвовавших в сражениях того времени. Наездник из Сёсдалы не был одиноким всадником! Он был лишь одним из нескольких представителей воинской знати, вернувшихся домой с грузом новых идей.

107 107 Igor GAVRITUKHIN Brooches of the Levice-Tokari Sub-Group This paper discusses brooches of the Levice- Tokari sub-group, which belong to the circle of interregional artefacts uniting the fifth century populations from Pannonia to the Black Sea and Don areas and demonstrating their connections as far as the Baltic Sea and Norwegian territory. 1. Definition Joahim Werner has distinguished the brooches in question as Prša-Levce type (Werner 1959: ). Peter Tomka has described the history of their studies in details (Tomka 2010: ), so I can only supplement his essay by mentioning several publications (Амброз 1971: 104; 1992: 73; 1994: 35; Tejral 1985: , 381, 383, 385; 1988: 278, 285, 286; 2012: 118, ; Айбабин 1990: 19, 60, 63; 1999: 268, табл. XXV, XXVII; Веймарн, Айбабин 1993: 182; Haralambieva 1990: 80; 1997: 19; Хараламбиева, Атанасов 1993: , 109; Bierbrauer 1995: 550, 568, 579; 2008: ; Kiss 1996: 59, 61, 64 65; Harhoiu 1998: ; Istvánovits, Kulcsár 1999: 88; Хайрединова 2002: 59 60, 78 80, рис. 18; Гавритухин 2004а: ; 2005: ; Gavritukhin 2012). There are detailed catalogues and maps (Heinrich 1990: Abb. 2; Гавритухин 1994: прил. 1; Bierbrauer 2008: 126, Abb. 17). This paper develops the points of my paper, published almost 20 years ago (Гавритухин 1994), against new materials 1 and observations (particularly Гавритухин 2004a: ; Gavritukhin 2012). The Levice-Tokari sub-group unites comparatively small (mainly cm long, but there also are 5.4 to about 9 cm long finds) cast brooches, ornamented by chip carving (Kerbschnitt), with simple design (combining triangles and quadrangles), extensions of the head-plate (one or three) and foot-plate (one or three), support for the rod of the spring, and solid catch-plate. The head-plate is triangular or segment-shaped, the foot-plate is sub-diamond. Extensions along the vertical axis are in the form of circular or faceted knobs, sometimes of complicated forms. Side extensions, if any, are knob-shaped or of simplified flattened forms. The material is mainly bronze, with few silver and gilt specimens; spring and needle are usually of iron. 2. Typology 2.1. The main series (fig. 1; ; 5) unites the finds with their main features very close to each other: triangular head-plate decorated with four incised triangles, diamond foot-plate decorated with four incised diamonds, and one projecting knob of simple form on the head- and foot-plates. They are mostly mm long, with a few deviations. As a rule, all the brooches have a rib on the back side of the plates, along the symmetry axis, sometimes pronounced (with sub-rectangular cross-section) and sometimes smoothed Cluster of variants Levce (corresponds to Hauptgruppe in: Heinrich 1990) with distinctive incised oval with pointed ends (Spitzovale) on the bow. Hopefully, the future studies will suggest variants singled out according to small differences in size and proportion of the plates, circular or cut form of the knob, and presence (or absence) of borders where the bow starts, which is not always understandable by the information in my possession. So far I have suggested variants only after the existence of elongated knob on the foot-plate and hollow bow. Variants not specified: Bački Monoštor (Бачки Моноштор; Bodog-Monostorszeg), grave 9 (1/1902) (fig ; 5. 17; Werner 1959: 429, no. 5, Abb. 5. 2; Kiss 1983: 101, no. 1; Vinski 1 I am indebted to Igor Bazhan, Eszter Istvánovits, Dagmar Jelínková, Valéria Kulcsár, Heino Neumayer, Karol Pieta, Erdmute Schultze, and Jaroslav Tejral who helped me a lot supplying information about the finds and sending copies of books that are hardly available in Russia.

108 108 Igor GAVRITUKHIN Fig. 1. Brooches of the main series. 1 Prša; 2, 3 Levce; 4 Velikaya Bakta; 5 Pobedim; 6 Hungary ; 7 Bojná; 8 Lužianky; 9, 10 Chotin; 11, 12 Jászberény; 13 Csongrad; 14, 15 Bački Monoštor; 16 Szilágy; 17 Carnuntum; 18 Steinbrunn; 19 Mitterhof; 20, 21 Bratei; 22 Bistrets; 23 Parndorf

109 Brooches of the Levice-Tokari Sub-Group 1987: Taf. VI. 1 2). Bojná, fortified settlement II (fig. 1. 7; 5. 7; Pieta 2007: 175, 179; obr. 3. 4; 2008: 468, 469; Abb. 6. 6); Karol Pieta kindly informs that three more brooches of the type were recently found in that site. Bratei, grave 2/1968 (fig ; 5. 21; Bârzu 1986: , fig ). Carnuntum (now Deutch-Altenburg) (fig ; 5. 2; Werner 1959: 429, no. 8а, Taf. VI. 10). Chernookovo (Чернооково) (fig , 5. 22; Хараламбиева, Атанасов 1993: 117, 11, табл. I. 2). Chotin (fig ; 5. 9) I am indebted to Karol Pieta for the information about this find, the drawing forwarded, and the permission to publish this artefact. Csongrád-Kenderföldek (Csongrád- Kaserne; fig , 5. 14), grave 133 the first publication cites two brooches of interest, though later we knew about three very similar specimens from this assemblage (Párducz 1963: 51, Abb. 9. 3, Taf. VIII ; Germanen 1987: 147, no. II.33.b); grave 58 or 85 caption to figures in the publication refers to grave 58, thought he description of the assemblage misses this brooch); as Istvan Bóna has inferred, it belonged to destroyed and never described grave 85 (Párducz 1959: , 339, Abb , Taf. XI. 5. a b; Bóna 1991: 248, no. D/8); upon my request, Eszter Istvánovits has kindly clarified the account with the Csongrádi Múzeum: no in grave 133 includes three fibulae of my present interest, and one more similar piece is indicated and a find without an assemblage; Werner possibly published a brooch listed above from Csongrád, with incorrect reference to some Budapest museum (Werner 1959: 428, no. 3). Hungary (published 1881, originated from the Kingdom of Hungary within the Austro-Hungarian empire) (fig. 1. 6, not on the map; Werner 1959, S. 429, 13, Taf. VI. 9); Kerch ( Керчь ; Kerč ) 2 (fig ; Kühn 1974: Taf ). Jászberény-Szőlő dülő, grave 1 (fig , 12; 5. 13; Párducz 1959: 318, Taf. XXI. 13, 14; XXII. 2, 3); Košice (Kassa, Kaschau) (fig ; Werner 1959: 429, no. 12; Bóna 1991: 248, D/6); Levce (Léva, Lewenz; fig. 5. 8), grave from 1899 (fig. 1. 2, 3, from the original, Carpathian Regional Museum, Uzhgorod, no. Б / 2, 3; Werner 1959: 429, no. 10а, Abb. 6; Bóna 1991: 248 [no. D/1]; 86, Abb. 33; Gavritukhin 2012: рис. 1. 4, 5); with a reference to the museum in Košice (Werner 1959: 429, no. 10b; 109 Taf. VI: 7); Lužianky (fig. 1. 8; 5. 6; Werner 1959: 429, no. 9, Taf. VI. 6; Točík 1962: , obr. 8. 2; Bóna 1991: , no. D/11; Pieta 1993: 78, оbr ). Mitterhof, grave 3 (fig , 5. 1; Heinrich 1990: 88, Abb. 7. 4, Taf. 2. 2; with clarification according to the original at the Museum of Natural History, Vienna, no ). Parndorf (fig , 5. 37; Nowak, Roth 1996: 729, Abb. 578). Pobedim (fig. 1. 5, 5. 5; Bialekova 1970: ; оbr. 1). Prša, grave 27 (from the age of the Avar khanate) (fig. 1. 1, 5. Točík, Drenko 1950: оbr. 99; Točík 1962: 202, obr ; Werner 1959: 428, 429, 11, Taf. VI. 5). Steinbrunn (fig , 5. 3; Mitscha-Märcheim 1966, S. 110, Abb. 2. 5). Southern or Eastern Crimea (not mapped; Платар 2004: 204, no. 17). Szekszárd-Plánk, grave 100 (from the age of the Avar khanate) (fig , 5. 15; Kiss 1996, S. 57, Abb. 9). Szilágy-Aranyodal (fig , 5. 16; Werner 1959: 428, 429, no. 4, Taf. VI. 8). Velikaya Bakta (Великая Бакта), grave 2 (fig. 1. 4, 5. 12; Черкун 1994: 93, рис. 5. 1; here I provide a drawing from the original, being acquainted to it by courtesy of the excavator Vyacheslav Kotigoroshko; Archaeological Museum at the University of Uzhgorod, no. 1532). Besides, the internet information concerning illegal finds mentions such brooches in: Borshchyov district (fig ); Boyarka (Боярка; fig ); Gorenichi (Гореничи; fig ); Kalinovka (Калиновка; fig ); Poltava district (fig ); Strel niki (Стрельники; рис ); Khmel nitskiy oblast (administrative area), two specimens (fig ); Chapayevka (Чапаевка; fig ); Chigirin district (fig ); two more specimens do not have other coordinates than simply the Ukraine (not on the map). Variant with elongated knob: Szekszárd-Plánk, grave 210 (fig , 14, 5. 15; Kiss 1996: 54, Abb. 5, 15. 1, 2). Variant with massive hollow bow: Bistrets (Бистрец; fig , 5. 23; Хараламбиева 1992: 137, no. 28, табл. IV. 5). Siscia (modern Sisak; fig , 5. 18; Simoni 1989: 112, 121, no. 26, T. 3. 7) Other clusters of variants combine varied specimens without oval on the bow. Variant with elongated knob: Viminacium (near modern Požarevac) (fig , 5. 19; Виминацjум 2 Werner has supplied the list of the brooches of type Levce at the Museum of Prehistory and Early History, Berlin, indicating their numbers (Werner 1959: 428, NN 1а-1f). However, after my request to the Museum, Heino Neumayer has kindly informed that such brooches are not in the collection under such numbers, with the only exception (see variant Fîntînele below).

110 110 Igor GAVRITUKHIN Fig. 2. Brooches of series Danube-Done, series Pannonia, and main series. 1 Alyoshki; 2, 7, 18 Ksizovo; 3, 4 Botoşani; 5 Veliko Tarnovo; 6 Rusensko; 8 Carnuntum; 9 Levél; 10 Halbturn; 11 Katzelsdorf; Szekszárd; 15 Viminacium; 16 Skalistoye; 17, 23 Kerch ; 19 Siscia ; 20 Chernookovo; 21 Dobrogea ; 22 Fîntînele

111 Brooches of the Levice-Tokari Sub-Group 1980: no ; Ivanišević, Kazanski, Mastykova 2006: 50, pl. 46, fig ). Variant with bow of triangular cross-section and small knob: Kerch ( Керчь, Kerč ) (fig , 5. 36; Айбабин 1990: рис ). Ksizovo (Ксизово) 19, area 2 (fig , 5. 34; from the original at the State Directorate for the Cultural Heritage Protection in Lipetsk Oblast, Lipetsk; Обломский 2011: pис ). Variant with narrow bow and solid foot-plate: Dobrogea (fig , 5. 24; Bârzu 1986: fig. 5. 5; Harhoiu 1998: Taf. LXXXIX.D). Besides, the internet information concerning illegal finds mentions one such brooch in the Ukraine (not on the map). Variant Fîntînele, less than 60 mm long, with cordons at the base of the bow and cut knob: Fîntînele, grave IV (fig , 5. 20; Crişan, Operanu 1991: ; fig. 2. 1; Harhoiu 1998: Taf. LXXXVIII.C. 18). Kerch ( Керчь, Kerč ) (fig , 5. 36; from the original at the Museum of Prehistory and Early History, Berlin, no. IIId.6015; Werner 1959: 428, no. 1а, Taf. VI. 12, Abb. 5. 1; Gavritukhin 2012: рис. 1. 1). Besides, the internet information mentions analogous brooch among illegal finds from the Ukraine (probably from the Kharkov oblast ; not on the map). With features not understandable from the available information to make a reliable attribution to a variant: Kerch ( Керчь, Kerč ) (fig ; Kühn 1974: Taf ). Rattenberg (fig ; 5. 38; Heymans 2004: 760, Abb. 1000). Most probably, within the same circle but of a particular variant: Skalistoye (Скалистое), vault 190 (fig , 5. 35; Веймарн, Айбабин 1993: 32, рис ; Хайрединова 2002: рис. 18: 5) Fragments not definable rather than within a series (although they could actually belong to another series, the main series is most probable if one takes into account the composition of finds in adjacent territories): Chotin (fig. 2. 9; 5. 9): I am grateful to Karol Pieta for this information, drawing forwarded, and the permission to publish the find; Katzelsdorf (fig , 5. 4; Tejral 2002: Abb. 2. 8) Series Danube-Don (fig ; 6. a e) unites the brooches decorated similarly to the main 111 series (more complicated ornamentation is still possible, like those of other series to be described below), but with larger head-plate to be closer to equilateral triangle, more complicated forms of knobs and / or side extensions on the plates, and bugger dimensions. The clusters of variants are singled out according to the form of the extensions at the tip of the foot-plate Cluster of variants Rusensko-Alyoshki is typical of trefoil extension. Details of each artefact considerably vary thus allowing its attribution to a specific variant: Alyoshki (Алёшки, modern Tsyuryupinsk) (fig. 2. 1, 6.14, drawn from the photograph in Galina Korzukhina s archive, the Institute of History of the Material Culture of the Russian Academy of Sciences archive, file 86). Ksizovo 19 (Ксизово 19), the area in between trenches 6 and 7 (fig. 2. 7, 6. 12; from the original at the State Directorate for the Cultural Heritage Protection in Lipetsk Oblast ; Обломский 2011: pис ; Gavritukhin 2012: рис. 2. 4). Rusensko, Ruse oblast (Русенско, област Русе) (fig. 2. 6, 6. 4; Haralambieva 1990: 80, Abb. 2. 4; Хараламбиева 1991: 36, обр. 7). Besides, the internet mentions a brooch of this circle among illegally discovered finds from the Sumy oblast (fig. 6. 9) Cluster of variants Sakhnovka has characteristic extension similar to elongated triangle or pointed oval form, sometimes with relief design resembling pentafoil; there are variants with different location of maximum width of the footplate, with or without side extension on the plate: Veliko Tarnovo (Велико Търново), Tsarevets hill (Царевец; fig. 2. 5, 6. 5; Haralambieva 1990: 80, Abb. 2. 5; Хараламбиева 1991: 36, рис. 9). 3 Besides, the internet mentions such brooches among illegal finds in: Zadonsk district (fig ); Kinburn Spit (Кинбурнская коса; fig ); Kharkov oblast (fig ); Sakhnovka (Сахновка; fig. 6. 8); Ukraine (not on the map) Particular variant of artefacts similar to the main series but with corrugation of the footplate and at the base of the elongated extension at the foot-plate. It is known with illegal finds pub- 3 According to the publication, this fragment was the largest one and had knob on the bow, similar to beast s head. Anna Haralambieva has supposed that the missing head-plate was semicircular and has related this find to radiate-headed brooches with knobbed head-plate. However, its main features are closer to other finds in the cluster under analysis; zoomorphic tip of the foot-plate is probably supplied by local artisan or as it was the publisher s interpretation.

112 112 Igor GAVRITUKHIN lished in the internet: Kharkov oblast (fig ); Khmel nitskiy oblast (fig. 6. 7). of variants are separated by the form of the tip of its foot-plate Variant Ksizovo is so far represented by the only artefact with the circular knob at the end of the foot-plate: Ksizovo (Ксизово) 19, area 5 (fig. 2. 2, 6. 12; from the original at the State Directorate for the Cultural Heritage Protection in the Lipetsk Oblast ; Gavritukhin 2012: рис. 2. 3) Fragments of a pair of brooches, with missing endings of the foot-plate: Botoşani-Dealul Cărămidăriei, grave 1 (fig. 2. 3, 4; 6. 6; Zaharia, Zaharia 1975: 205; fig , 2; 14. 1, 2) Series Pannonia (fig , 6. f) unites the brooches with vertically elongated head-plate, decorated in the same way as the main series, or with complicated combinations of triangles; another difference from the main series is their bigger size, complicated form of extensions at the foot-plate, and sometimes side extensions on the plates. The finds have considerably different features allowing to single out particular variants (or, taking into account different tip of the foot-plate and analogy with series Danube-Don and Black Sea, to the cluster of variants): Carnuntum (modern Deutch-Altenburg) (fig. 2. 8; 6. 1; Werner 1959: 429, no. 8b, Taf. VI. 11; Germanen 1987: 330, 331, no. VII.1.g). Levél, site 234 (fig. 2. 9; 6. 3; Tomka 2010: , Abb. 4). Halbturn (fig , 6. 2; Holl 1983: 46, Taf. 1. 2; clarified drawing from the original at the Museum of Natural History, Vienna, no ; Germanen 1987: 338, 339, no. VII.24; Gavritukhin 2012: fig. 1. 3) Specific find has particular form against the background of the brooches under analysis, though generally similar to some of them. Only new finds can allow the answer if it forms specific variant (or cluster of variants) within established series or particular series: Kerch ( Керчь, Kerč ) (fig , 6. g, from the original at the Museum of Prehistory and Early History, Berlin, no. IIId.6382; Kühn 1974: Taf ; Gavritukhin 2012: рис. 1. 2) Series Black Sea (fig , 7. a d) unites brooches with segment-like (semicircular) head-plate and more complicated decoration than the mains series, often with more complicated forms of the extension at the head-plate, side extensions at the plates, and larger size. The clusters Cluster of variants Tokari-Sakharnaya Golovka is characteristic of trefoil tip of the footplate. Variants are determined according to the particulars of the ornamentation. The finds of variant Tokari have the head-plate ornamented like the main series (with consideration of the particulars of the contour of the plate), and elongated triangles on the foot-plate: Tokari (Токари) (fig. 3. 9; 7. 3, drawing by Rostislav Terpilovskiy; Гавритухин 1994: 39, no. 1 [the paper referred to was not published], 40, рис. 2. 7). The finds of variant Sakharnaya Golovka have the head-plate ornamented with various combinations of six triangles: Ksizovo (Ксизово) 8 (fig. 3. 6; 7. 4; from the original at the State Directorate for the Cultural Heritage Protection in the Lipetsk Oblast, Lipetsk; Обломский 2011: fig ; Gavritukhin 2012: fig. 2. 2). Ksizovo 19 (Ксизово 19), area 3 (fig. 3. 5; 7. 4; from the original at the State Directorate for the Cultural Heritage Protection in the Lipetsk Oblast ; Обломский 2011: рис ; Gavritukhin 2012: fig. 2. 1). Sakharnaya Golovka (Сахарная головка), grave 12 (with two undercuts no. 1) (fig. 3. 7; 7. 6, from the original at the National Preserve of Tauric Chersonesos (Sevastopol), no. 1/36276 I am indebted to Larisa Golofast for her help; Борисова 1959: 182, Табл. VI. 3; Айбабин 1990: рис. 14: 6; Гавритухин 1994: 39, 4, рис. 2. 9; 2004a, рис ; Хайрединова 2002: рис ). Besides, the internet informs about analogous brooch among the illegal finds from the Ukraine (not on the map) Cluster of variants Kerch-Kopachintsy has typical simple extensions, ornamentation of three embossed triangles on the head-plate, and eight incised triangles on the foot-plate. These variants differ by the form of extensions and presence or absence of extensions at the sides of the plate: Kerch (Керчь, Kerč), vault 66/1905 (fig ; 7. 8, drawing by Anatoliy Ambroz; Гавритухин 1994: 39, no. 5, рис ; most likely, it is published in: Калитинский 1928: табл. XXXIV. 27, and re-published from there). Besides, the internet mentions such brooches among illegal finds in: Kopachintsy (Копачинцы; fig. 7. 1); Priluki district (fig. 7. 2) Specific variant Chersonesos (variant Zyuk according to: Гавритухин 2004a: ; it is renamed due to the terminology at: Bierbrauer

113 Brooches of the Levice-Tokari Sub-Group 113 Fig. 3. Brooches of series Black Sea, series Balkans, Kerch artefact, and hybrid pieces. 1, 2, 4, 8 Chersonesos; 3 Zyuk; 5, 7 Ksizovo; 6 Sakharnaya Golovka; 9 Tokari; 10 Kerch; 11 Vyrbitsa; 12 north-eastern Bulgaria ; 13 Dragoevo; 14 Kerch ; 15 Szurdokpüspöki; 16 Záhony

114 114 Igor GAVRITUKHIN 2008: 126, Abb. 17) is characteristic of long end of the foot-plate, ornamentation of three convex triangles on the head-plate, and seven incised triangles on the foot-plate. The form of the knob on the headplate is close to hemisphere and side extensions to rectangle. On the reverse side there was vertical rib similar to the finds of the main series to separate the variant under study from those of the Black Sea and other series but the main one: Chersonesos (Херсонес, modern Sevastopol; fig. 7. 5), vault 3/1891 (fig. 3. 1, 2; Зубарь 1982: 59, рис ; Айбабин 1990: рис , 4); the find of 1887 (fig. 3. 8, from the original at the State Hermitage Museum, Byzantine Department, no. Х-679; Гавритухин 2004a: 213, рис ; Залеская 2006: no. 181); the collection of the State Historical Museum, Moscow (fig. 3. 4, drawing by Anatoliy Ambroz; Гавритухин 1994: 39, no. 3, рис ). Zyuk (promontory on the southern coast of the Azov Sea close to modern Kurortnoye village; мыс Зюк), town area, grid А-7 (fig. 3. 3, 7. 7, from the original at the Kerch Historical and Cultural Preserve; Масленников 1992: 152, рис ; Гавритухин 1994: 39, no. 2, рис ) Series Balkans (fig , 12; 7. е) unites brooches with the head-plate similar to horizontally-elongated triangle, of demonstrative small size (less than 6 cm long). These artefacts have considerably different details allowing us to put every of these brooches to particular variant: north-eastern Bulgaria (fig , not mapped, from the original, acquired by the Archaeological Museum in Varna; I am indebted to the museum staff for their help with working with this find and the permission to publish this find). Shumensko, Shumen oblast (Шуменско, област Шумен) (fig. 7. 9, the find similar to the find from Varbitsa; Хараламбиева, Атанасов 1993: 107). Varbitsa (Върбица) (fig ; 7. 10; Хараламбиева, Атанасов 1993: 107, 115, 1; табл. II. 3). 3. The chronology and the cultural-historical context It is almost accepted so far that the main series of brooches of interest dates back to phase D2/D3, possibly including phase D3. 4 They are examined within the styles of the most simple geometrised chip carving design that appeared with barbarian artefacts influenced by Roman specimens. In phase D2/D3 and possibly phase D3, there are several remarkable types of fly brooches (Zikadenfibeln) (fig. 4A. 18, 21) and buckles with decoration analogous to that of the brooches of interest (fig. 4А. 5), which occur in assemblages with the brooches of the main series. Other finds from these assemblages with wider chronology do not contradict to the above-suggested date as well (fig. 4А. 1 8, 12 21). Distribution area of the main series of brooches (fig. 5) enveloped the territories within the Hunnic realm or under its control in the age of Attila. 5 The finds form almost a continuous tract except for mountainous areas with no sites from the period of interest and the steppe of Huns and other nomads that did not wear brooches. Areas of later types of brooches, including some variants within the Levice-Tokari sub-group, but not of the main series (fig. 6, 7), are much more compact and show specific connections rather than wavy-like distribution in the homogenous environment. These observations make an argument to date the appearance of the main series of brooches of the Levice-Tokari sub-group to the early period of phase D2/D3, i. e. before the battle of Nedao (AD 453/56), when the single space of political and cultural of the Hunnic realm disintegrated. Bow decoration of the cluster of variants Levice is typical of incised oval with pointed ends. Similar but longer ovals on the bow are known among some brooches of type Wiesbaden (fig. 4В. 3, 4), from the first half of the fifth century, distributed far to the north-west than the area of the Levice-Tokari subgroup (Werner 1982; Mahajewski 2008: Abb. 18). On the edge with the area of type Wiesbaden are the brooches of type Hácz-Béndekpuszta (Attila Kiss Béndekpuszta), decorated with incised design with pointed ends (however, on the head-plate) and the form of foot-plate like that of type Wiesbaden (such as fig. 4В. 2; Tejral 2008: 258; Abb. 4). It is also the area of the main concentration of the finds of the main series of the Levice-Tokari sub-group (fig. 5). Outside of the Empire, this area from the Roman period was occupied by groupings of the Suebi and Quadi, a part of which rest in place after their relatives moved to the west, finally forming the Suebian kingdom in the north-western area of the Pyrenean peninsula. It is worth mentioning that the finds with incision on bow of type Wiesbaden originate from the 4 Henceforth I refer to phases according to Jaroslav Tejral except for a few cases especially underlined. 5 Although I have made particular researches of purely archaeological and historical reasons behind this chronology, their results are united here to save the volume of this paper and for convenience of comparison.

115 Brooches of the Levice-Tokari Sub-Group 115 Fig. 4А. Artefacts uncovered within the assemblages containing brooches of the Levice-Tokari sub-group. 1 6 Bački Monoštor, grave 9; 7, 8 Skalistoye, vault 190; 9 11 Sakharnaya Golovka, grave 12; Mitterhof, grave 3; Fîntînele, grave IV; 18 Velikaya Bakta, grave 2; Levce, 1899 grave Fig. 4В. Analogies to the elements of the brooches of the Levice-Tokari sub-group. 1 Vrchoslavice; 2 Hácz-Béndekpuszta; 3 Foss; 4 Dębczino; 5 Herpes; 6 Bulles; 7 Tiszalök; 8 Mád; 9 Kerch; 10 Castelbolognese; 11 Zemun; 7 10 details

116 116 Igor GAVRITUKHIN Fig. 5. The finds of the main series. Imprecise locations are marked as: I within district, radius ca km; II within oblast, radius ca km; III seller s connection of the find to a popular archaeological site. a c cluster of variants Levce (a variants not specified; b with elongated knob; c with hollow bow); d h no oval on the bow (d with elongated knob; e with bow of triangular cross-section; f with narrow bow and massive foot-plate; g variant Fîntînele; h the find from Skalistoye); i fragment or artefact with unclear details. 1 Mitterhof; 2 Carnuntum; 3 Steinbrunn; 4 Katzelsdorf; 5 Pobedim; 6 Lužianky; 7 Bojná; 8 Levce; 9 Chotin; 10 Prša; 11 Košice; 12 Velikaya Bakta; 13 Jászberény; 14 Csongrad; 15 Szekszárd; 16 Szilágy; 17 Bački Monoštor; 18 Siscia ; 19 Viminacium; 20 Fîntînele; 21 Bratei; 22 Chernookovo; 23 Bistrets; 24 Dobrogea ; 25 Borshchyov district; 26 Khmel nitskiy oblast ; 27 Boyarka; 28 Gorenichi; 29 Chapayevka; 30 Chegirin district; 31 Kalinovka; 32 Strel niki; 33 Poltava district; 34 Ksizovo; 35 Skalistoye; 36 Kerch ; 37 Parndorf, 38 Rattenburg

117 Brooches of the Levice-Tokari Sub-Group 117 northern areas, from Norway to Polish Pomerania (Werner 1982: ; Beilage 2, no. 21, 24). Traditional connections of these areas have been discussed many time, though brooches of other types, also with incision on the bow, concentrates in southern Scandinavia (for example: Bitner- Wróblewska 2001: pl. LII. 5, LI. 2, XLVI. 3, XLIV. 5). This circle was probably related with the find from Vrchoslavice in Moravia (fig. 4В. 1; Tejral 2008: Abb ), similar to the finds of type Wiesbaden, though following another tradition (with vertical stand for spring axle). Archaeological finds Scandinavian-Danubian contacts in the period of interest have been addressed many times on account of archaeological finds, though in some cases written sources allow us to relate some of these finds with the Heruli s activities. In the second half of the fifth century, the Heruli were among the leading political forces north of the middle Danube area, and obviously included a part of Suebian and other groupings. The Heruli came to that area most likely about the second quarter of the fifth century, and they probably revived contacts with their remote homeland in Scandinavia about the same time (Gavritukhin 2011: 56 57). Thus, the brooches of the Levice-Tokari subgroup appeared in the early phase D2/D3 (about the second quarter of the fifth century) in relation to the Suebian-Herulian environment in the Danube area (cf.: Heinrich 1990: 96 97; Tomka 2010: 479, 485), which was subordinated to the Huns in that period. Such brooches in the middle Tisza area were discovered in Sarmatian graves, thus agreeing with traditional context of Sarmatian-Suebian/Quadian connections, which had been strong enough since the Roman period onwards. The period of Hunnic dominance over the large zone of non-steppe populations was accompanied with the shaping of around-hunnic elites sharing common culture to be reflected both in the equipment of mounted warriors (basically against the background of eastern samples) and in female attire (basically with Eastern Germanic, Alanic, and late antique components). A series of populations headed by these elites distributed brooches of the main series of the Levice-Tokari sub-group. These brooches were still used (and possibly produced) after the collapse of the Hunnic realm. The date of grave 9 in Bački Monoštor (with the brooch of cluster of variants Levice) is based mainly on the buckle chronology (fig. 4А. 5). Its most close analogy (of better quality as the publication suggests) appeared in Székely with twoplate brooches of series Székely, one of the reference assemblages in phase D2/D3 (Tejral 1997: , Abb ; for series Székely see: Гавритухин, Казанский 2006: 297; Gavritukhin, Kazanski 2010: 87 88). Unfortunately, the publication does not disclose the function and details of one find from grave 9 in Bački Monoštor (fig. 4А. 3). It is similar to vertical knob on the head-plate of a large two-plate brooch with appliqués (for example: fig. 4В. 7). Following Volker Bierbrauer, I consider these artefacts to be demonstrative for phase D3 (Гавритухин 1994: 34 36, with earlier bibliography and discussion; Bierbrauer 1995; 2008). If my interpretation of this publication is correct, grave 9 in Bački Monoštor should be dated to the late phase D2/D3, or partly synchronous to phase D3. The chronology of Szekszárd-Palánk is based on brooches of the cluster of variants Levice and brooch from grave 217, typical of phase D3. This compact cemetery definitely existed for a short time, and the correlation of chronological indicators suggests the crossing of phases D2/D3 and D3. By all appearance, Attila Kiss has correctly related this site, due to its time and place, to Theodemir s Goths, suggesting its date as AD 456/ (Kiss 1996). According to publications, brooches from Bački Monoštor and Szekszárd-Palánk are poorly preserved and some parts are missing (fig , 15; ), a possible evidence of their long existence. Elongated knob of the pair of brooches from Szekszárd-Palánk is most likely a result of re-working of original samples. Hollow bow could possibly be a marker of relatively late specimens. This feature is known with only two brooches of cluster of variants Levice, though it often occurs among the specimens of series later than the main one (for example fig. 3, 4 8, 14). Large brooch of the main series from vault 190 of Skalistoye was accompanied with a small belt buckle (fig. 4А. 8), described by Elzara Khairedinova as that of Byzantine circle, with wire frame and tongue, of [Aleksandr Aibabin s] variant 1 and dated to the second half of the fifth century after Aibabin (Хайрединова 2002: 55; Айбабин 1990: 36), probably because of the chronology of the brooch. Khairedinova has attributed analogous shoe buckles to variant 1 (the earliest) of type 2 from the first half or entire sixth century (Хайрединова 2003: 127). In the assemblages important for chronological analysis, these

118 118 Igor GAVRITUKHIN Fig. 6. The finds of series Danube-Don, series Pannonia, and a similar artefact. Imprecise locations are marked as: I within district, radius ca km II within oblast, radius ca km III seller s connection of the find to a popular archaeological site. a е series Danube-Don (a cluster of variants Rusensko-Alyoshki; b cluster of variants Sakhnovka; c particular variant; d variant Ksizovo; e the find from Botoşani); f Pannonia series; g specimen similar to the above. 1 Carnuntum; 2 Halbturn; 3 Levél; 4 Rusensko; 5 Veliko Tarnovo; 6 Botoşani; 7 Khmel nitskiy oblast ; 8 Sakhnovko; 9 Sumy oblast ; 10 Kharkov oblast ; 11 Zadonsk district; 12 Ksizovo; 13 Kinburn spit; 14 Alyoshki; 15 Kerch

119 Brooches of the Levice-Tokari Sub-Group 119 buckles regularly appear with small belt-ends (like that on fig. 4А. 9, see below), Byzantine buckles with sunken parallel lines on the panel (type V in: Kazanski 1994), 6 and their non-ornamented analogies. It supports the many researchers chronology of Skalistoye vault 190 as the second half of the fifth century. All the mentioned comparatively late assemblages with brooches of the main series of the Levice-Tokari sub-group are located outside the underlined above zone of these brooches origin. It certainly does not imply that such brooches were not worn in the north-western part of their area after the battle of Nedao or that they appeared in other areas only after that battle. By all appearance, brooches of the cluster of variants Levice (and possibly of some other artefacts of the main series, like variant Fîntînele) were widely distributed as late as the age of Attila, though the tradition of their production did not continue everywhere. On the north and east of the Carpathian basin, brooches of the main series of the Levice-Tokari sub-group were replaced by other types of brooches rather than their own derivatives. Following Werner, Kiss has suggested a typology line from Levice type of brooches to types Dunapentele/Gyula and Miszla (Kiss 1996: 64 65, Abb. 14) and similar artefacts, which I consider very schematic. More likely, these types of brooches document various types of adaptation of Roman chip carving in the environment of the middle Danube barbarians other than the strict sequence and continual development. On the north-east of Carpathian Hollow, there are brooches which probably developed under the influence of the Levice-Tokari sub-group. They include brooches of type Záhony (as on fig ; generalizing drawing from the pair of brooches at the Jósa András Museum, Nyíregyháza). For understanding of the tradition of their making, there is a demonstrative technology of fastening of the spring with two stands on either side of the headplate, similar to many types of Danubian brooches from phase D3. Ornamentation features of brooches of type Záhony are similar to specimens of series Black Sea of the Levice-Tokari sub-group (for example: fig ), probably reflecting selective connections of the upper Tisza area with the lands far to the east, as it also documented by the finds of brooches of type Záhony in the Taman peninsula (according to the sellers). This similarity was the reason why I attributed the find from the Taman peninsula to the Levice-Tokari sub-group in my paper of 1994, though later on I reconsidered that attribution (Гавритухин 2004a: 213). Most likely, the pair of brooches from Szurdokpüspöki (fig ) belongs to the same context, though only a front photograph is published so far (Bácsmegi, Guba 2007; Tejral 2008: Abb ). Independent development of brooches of the Levice-Tokari sub-group on the west of their area is traced only with a very compact group of sites on the north-western extremity of Pannonia, as a very few finds attributed to series Pannonia (fig , 6. f). Their chronology has been determined by the later typological position in comparison with the main series of brooches: the dimension bigger than 7 cm (as Tomka has pointed out), complicated form of vertical extensions, side extensions on some specimens, and complicated ornamentation of plates. As for the stylistic analogies, let us mention a knob on the foot-plate, which has the form of trefoil or half-opened blossom, also occurring among the cluster of variants Rusensko-Alyoshki of the series Danube-Don and cluster of variants Tokari-Sakharnaya Golovka within the series Black Sea (fig. 2. 1, 6 8; , 9). The researchers have already mentioned that analogous design of some two-plate brooches with appliqués and a pair of brooches from Zemun with simple chip carving (fig. 4В. 7 11; Bierbrauer 1995, S. 550; Abb. 2; 6. 4, 5; 7. 1, 2; 8. 1, 2), i. e. artefacts from phase D2/ D3 or phase D3 except for its latest period. This and other innovations recorded in series Pannonia are also observable in some finds of other series which have clear differences so that is the case of different series related with close but developing separately tradition. Although Bierbrauer s term open space (offene Räume) is good enough, 6 Aleksander Aibabin has dated the buckles with parallel lines on the panel to the first half of the sixth century, though Michel Kazanski has considered them to appear in the middle or the second half of the fifth century and to exist to the mid-sixth century (Айбабин 1990: 36; 1999: 269; Kazanski 1994: 163, 164; 2002: 146; Казанский 2001: 46). The reasons for their early date are related to the assemblages of Dyurso of Aleksandr Dmitriyev s phase 2 of this cemetery (the first and second thirds of the sixth century), though Kazanski has united them with phase 1, synchronous to the Central European phase D3 (Казанский 2001; Kazanski 2002). The early chronology is also supported by the materials from burial 1 in vault 3 in Karshi-Bair, having the set of belt fittings of interest and a pair of fly brooches similar to the example on fig. 4А. 18 (Ушаков 2010: pис. 80).

120 120 Igor GAVRITUKHIN I consider that the formation and the apogee of this phenomenon are related to the Hunnic period. The battle of Nedao marked the limit after which it did not actually exist, though its traces survived with the former unity of tradition and selective ties (cf.: Гавритухин 1994: 36 37). The collection of materials makes often interpretation of this space as the Eastern Germanic circle of cultures more and more conventional. Series Pannonia probably did not last for long. The problem of the reasons behind this phenomenon (it is worth remarking that on the left bank of the Danube, where Suebian groupings continued under the dominance of the Heruli, there were no brooches of series Pannonia) is a topic of particular discussion, related to detailed analysis of the historical situation in the middle Danube area after the death of Attila and the battle of Nedao. Here I will restrict myself simply to the statement of it. Especial investigation should discover if there was a connection between the brooches of Bulles- Lauriacum type, decorated with chip-carved triangles (as on fig. 4В. 6; Koch 1998, pt. II. 1), with the same tradition as brooches of the Levice-Tokari sub-group, 7 or the ornamentation of the former appeared as another line of reshaping of late antique samples. Another picture of the development of the brooches of the Levice-Tokari sub-group appeared in the east of their area, that is in the zone related to populations still under Hunnic protectorate in the battle of Nedao and later. The area of series Danube-Don is almost the same as the zone of eastern finds of the main series of the brooches except for the Crimea (cf. fig. 5 and 6. a d). These brooches are related to different populations. There also are artefacts of series Danube- Don discovered outside of archaeological contexts in the lower Danube and north-western Black Sea areas, but it is too early to discuss their cultural attribution. In the eastern Carpathians, a pair of brooches of interest are reliably related with the site (Botoşani; fig. 6. 6), which I have attributed to the circle of post-chernyakhov antiquities in that area. Its final stage was in the second half of the fifth century because of the appearance of the Prague culture in that area and the migration of its people (Slavs) to the lower Danube in the early sixth century (Гавритухин 2000; 2009: 13; with extra bibliography). The finds in the tract from the Dniester to the Severskiy Donets form a part of the Prague and Pen kovka cultures area that replaced there the Chernyakhov culture. Brooches of the latter culture are presented there by many hundreds of artefacts, including none of our present interest, thus indicating that the brooches of the Levice-Tokari subgroup occurred later. The interpretation of the fifth and sixth century Slavic attire as broochless is disproved by increasing number of finds and their detailed analysis. Recent studies have discovered artefacts suggesting local series and local production of brooches somewhere in the Pen kovka culture area as late as the late fifth century and relating them to the Chernyakhov tradition adopted in the new environment (Гавритухин 2007; 2010b; 2011). However, in the Prague culture area reliable local series occurred in the seventh century only (Гавритухин 1991; 2010а: 53 55); the attire of the Prague people in the eastern Carpathians was not brooch-less yet in the earlier period. Similar to the Pen kovka culture area, there the Slavs assimilated the population bearing the tradition originating from the Chernyakhov culture, though the development of the southern Prague culture area was greatly influenced by the contacts with Byzantium, which borders were approached by the Slavs as late as 520s. In the Don area, the brooches in question are undoubtedly related with the heterogeneous sites of type Chertovitskoye-Zamatino, except for their earliest horizons. The upper chronological limit of these sites is indicated by the artefacts with analogies in the Danube, among the antiquities of phase D3 (Гавритухин 2004b; Гавритухин, Обломский 2006; Обломский 2011); no reliable later goods are discovered there so far, despite of large-scale field researches. Specific picture is uncovered by the finds of series Black Sea (fig. 7. a d). Variant Tokari, the closets to the circle of prototypes (artefacts of the main series), is represented by eponymous find in the forest-steppe area on the left bank of the Dnieper. This place (fig. 7. 3) is located almost in the middle of the tract stretching from the upper Dniester area to the upper Don area and supplied four typologically similar but later brooches of variant 7 Although earlier I attributed the find from Herpes to this sub-group, recently I have got acquainted with the photograph from the British Museum (fig. 4В. 5) suggesting its interpretation within the early Merovingian circle to reflect the influence from cultures far to the east, as it has been described by colleagues many times.

121 Brooches of the Levice-Tokari Sub-Group 121 Fig. 7. The finds of the Black Sea and Balkan series. Imprecise locations are marked as: I within district, radius ca km. a d series Black Sea (a, b variants Tokari and Sakharnaya Golovka within the cluster of variants Tokari-Sakharnaya Golovka; c cluster of variants Kerch-Kopachintsy; d variant Chersonesos); е Balkan series. 1 Kopachintsy; 2 Priluki district; 3 Tokari; 4 Ksizovo; 5 Chersonesos; 6 Sakharnaya Golovka; 7 Zyuk; 8 Kerch; 9 Shumensko; 10 Vyrbitsa Sakharnaya Golovka and cluster of variants Kerch- Kopachintsy, occurring in the Crimea only twice. The chronologically latest variant Chersonesos has been recorded only in the Crimea so far, but there are at least four finds. It allows the suggesting that the beginning of the development of series Black Sea is related to the forest-steppe area of the Eastern Europe, where active re-working of the main series is also marked by series Danube- Don. Thence (probably via the upper Don area)

122 122 Igor GAVRITUKHIN the samples of the series Black Sea came to the Crimea, which was reflected by the finds of variant Sakharnaya Golovka. Cluster of variants Kerch- Kopachintsy documents the development of the same tradition with inclusion of Crimean elements (see on variant Chersonesos below). The isolation of variant Chersonesos demonstrates the gap in contacts between Crimean cultures and those in the forest-steppe area of Eastern Europe. Separate development of brooch production has been recorded in the forest-steppe zone from the last decades of the fifth century onwards, including local adaptations of Black Sea artefacts particularly presented by series Novoselitsa (Гавритухин 2007). An important reference point for the chronology of series Black Sea is grave 12 in Sakharnaya Golovka, with burial 1 containing iron two-piece brooch with scroll on the end of the solid catch-plate (fig. 4А. 11). In the Danube area, such brooches are united into type Viminacium from the second third of the fifth to the early sixth century (Ivanišević, Kazanski, Mastykova 2006: 17, with bibliography). 8 Burial 2 in grave 12 in Sakharnaya Golovka contained small strap-end (fig. 4А. 9). Khairedinova has convincingly dated such end-pieces from the second half of the fifth to the sixth century according to the Crimean materials (Хайрединова 2003: 132; for the reference assemblages see above, in the discussion of the chronology of vault 190 in Skalistoye). I can add to her arguments that such belt-ends appeared in Merovingian cemeteries in Northern Gallia in phase M1 from AD 470/80 520/30 and existed there to phase MR1, i. e. in 600/10 630/40 (type 199 according to: Legoux, Périn, Vallet 2004). The chronology of the given assemblage within the second half of the fifth or early sixth century does not contradict to the buckle discovered there (fig. 4А. 10). Specimens of variant Chersonesos stands out of other brooches of series Black Sea because of the long relief extension at the end of the foot-plate (fig , 8). This variant and cluster of variants Kerch-Kopchintsy (fig ) are also typical of the head-plate with embossed rather than incised ornamental triangles. The closest analogies in Eastern Europe to these elements and their combinations occur on local re-works of earlier variants of the brooches with the foot-plate decorated with concentric diamonds (type Gurzuf according to Kühn 1974, e. g.: Айбабин 1990: pис ), which are not earlier than their prototypes from phase D3 (Гавритухин, Казанский 2006: ; Gavritukhin, Kazanski 2010: ). A brooch of variant Chersonesos originates from the settlement layer on cape Zyuk in the eastern Crimea. Aleksandr Maslennikov has convincingly dated the collapse of this settlement to the second quarter of the sixth century (Масленников 1992). Quite probably, it happened in result of unrest caused by the activities of Hunnic prince Gordas, who controlled that area, or the Byzantine annexation of Bosporos, i. e. the events of the late 520s and 530s. The turn of 530s is also documented by the observations on the evolution of the Bosporan female costume. Its cardinal change is comparable with the appearance of Byzantine troops in Bosporos, which probably included the Germans with their families, though the artefacts originating from the Hunnic age had predominated in Bosporos before (Гавритухин, Казанский 2006; Gavritukhin, Kazanski 2010; Гавритухин 2011; these studies include a review of the problem and bibliography). Similar picture is documented in the lower Danube area. The finds of series Danube-Don indicate a connection to the strip of Eastern European sites stretching to the east of the Carpathians as far as the Don (fig. 6. a, b). Typologically latest specimens form local Balkan series known only in north-eastern Bulgaria (fig , 12; 7. e). The same area includes the find from Dragoevo (рис ; Хараламбиева, Атанасов 1993, табл. II: 6): a hybrid brooch, which small size, form and ornamentation of the head-plate are similar to the specimens of series Balkans, and form and ornamentation of the foot-plate to other types (cf. e. g. fig ). The analysis of typological lines and areas known so far suggests the most possible chronology of type Chersonesos in the period when the connections of the Crimea and the forest-steppe area of Eastern Europe interrupted, though the Hunnic tradition still continued, i. e. in the last decades of the fifth and thirst third of the sixth century. Similar or possibly a little bit earlier date is acceptable for brooches of series Balkans and two fibulae from 8 It is necessary to mention that another distribution peak of these brooches was in the period of the Saltov culture (type 7 at: Дмитирев 1982). The relation between these groups of finds is not clear, though if it existed, we do not know the transition link between them in the sixth and seventh century (for some considerations concerning the brooches with scroll on the tip of solid catch-plate see: Гавритухин 2008).

123 Brooches of the Levice-Tokari Sub-Group Botoşani (рис. 2. 3, 4), comparable with variant Chersonesos by the proportion of the foot-plate (its widened section is closer to the bow) and the side 123 knobs. These specimens are local forms of the final stage of the evolution of the brooches of the Levice-Tokari sub-group. Bibliography Bârzu L. Monumente germanice descoperite la Bratej, Jud. Sibiu // Studii şi cercetări de istorie veche şi archeologie / 1. Bialekova D. Nález spony z doby sťahovania národov v Pobedime // Študijné zvesti AÚ SAV Bierbrauer V. Das Frauengrab von Castelbolognese in der Romagna (Italien) Zur chronologischen, ethnischen und historischen Auswertbarkeit des ostgermanischen Fundstoffs des 5. Jahrhunderts in Südosteuropa und Italien // Jahrbuch des RGZM (1991). Bierbrauer V. Ethnos und Mobilität im 5. Jahrhundert aus archäologische Sicht: von Kaukasus bis Niederösterreich. München, Bitner-Wróblewska A. From Samland to Rogaland. East-West connections in the Baltic basin during the Early Migration Period. Warszawa, Bóna I. Das Hunnenreich. Budapest; Stittgart, Crişan I. H., Operanu C. Un mormînt din secolul V e. n. de la Fîntînele // Ephemeris napocensis I. Gavritukhin I. Cut Glass Beakers within the Context of Studies in the Connections between the South of Eastern Europe and Scandinavia in the Late Period of Roman Influence and the Great Migration Period // Inter Ambo Maria. Contacts between Scandinavia and the Crimea in the Roman Period. Kristinsand; Simfiropol, Gavritukhin I. Brooches of the Levice-Tokari Sub-Group // Inter Ambo Maria. Northern Barbarians from Scandinavia towards the Black Sea (Abstracts). Simferopol, Gavritukhin I. Kazanski M. Bosporus, the Tetraxite Goths, and the Northern Caucasus Region during the Second Half of the Fifth and the Sixth Centuries // Neglected Barbarians. Turnhout, Germanen, Hunnen und Awaren. Schätze der Völkerwanderungszeit. Ausstellungkatalog des Germanischen Nationalmuseums. Nürnberg, Haralambieva A. Archäologische Spuren der Goten südlich der unteren Donau aus dem 5. Jh. n. Chr. // AAustr Haralambieva A. Goten, Hunnen und Awaren in Nordostbulgarien (Süddobrudža und Ludogorie) // Miscellanea Bulgarica. 11. Wien, Harhoiu R. Die frühe Völkerwanderungszeit in Rumänien. Bukarest, Hеinrich A. Ein völkerwanderungszeitliches Gräberfeld bei Mitterhof, GB Laa an der Thaya, Niederöstrreich // AAustr Heymans H. KG Rattenberg // Fundberichte aus Österreich (2003). Holl A. Zwei völkerwanderungszeitliche Fibeln aus dem Burgenland // Annalen des Naturhistorischen Museums in Wien. 85/A. Wien, Istvánovits E., Kulcsár V. Sarmatian and Germanic People at the Upper Tisza Region and South Alföld at the Beginning of the Migration Period // L Occident romain et l Europe centrale au debut des Grandes Migrations / Spisy AÚ AV ČR. 13. Brno, Ivanišević V., Kazanski M., Mastykova A. Les nécropoles de Viminacium à l époque des Grandes migrations. Paris, Kazanski M. Les plaques-boucles méditerranéennes des Ve VIe siècles // Archéologie Médiévale XXIV. Kazanski M. Die Chronologie der Anfangsphase des Gräberfeld von Djurso // Probleme der frühen Merovingerzeit im Mitteldonauraum / Spisy AÚ AV ČR. 19. Brno, Kiss A. Die Skiren im Karpatenbecken, ihre Wohnsitze und ihre materriale Hinterlassenschft // AAH Kiss A. Das Gräberfeld von Szekszárd-Palánk aus der zweiten Hälfe des 5. Jahrhudert und der ostgotische Fundstoff in Pannonien // Zalai Múzeum Koch А. Bügelfibeln der Merowingerzeit im westlichen Frankenreich. Bonn, 1998.

124 124 Igor GAVRITUKHIN Kühn H. Die germanischen Bügelfibeln der Völkerwanderungszeit. Süddeutschland. Graz, Legoux R., Périn P., Vallet F. Chronologie normalisée du mobilier funéraire mérovingen entre Manche et Lorraine. Paris, Mahajewski H. Südzone des Ostseebeckens und Elbkulturkreis in der späten römischen Kaiserzeit und in der frühen Stufe der Völkerwanderungszeit // The Turbulent Epoch. New materials from the Late Roman Period and the Migration Period. II. Lublin, Mitscha-Märheim H. Das langobardische Gräberfeld von Steinbrunn und die völkervanderungszeitliche Besiedlung des Ortsgebietes // Wissenschaftliche Arbeiten aus dem Burgenland Nowak H., Roth H. KG Parndorf // Fundberichte aus Österreich (1995). Párducz M. Archäologische Beiträge zur Geschichte der Hunnenzeit in Ungarn // AAH XI. Párducz M. Die ethnischen Probleme der Hunnenzeit in Ungarn. Budapest, Pieta K. Osídlenie z doby rímskej a stahovania národov v Nitre // Nitra: Príspevky k najstarsim dejinám. Nitra, Pieta K. Gradiská Bojná II a Bojná III. Významné sídlo z doby st ahovania národov a opevnia z 9. storočia // Bojná. Hospodarárske a politické centrum Nitrianskeho kniežatstva. Nitra, Pieta K. Höhensiedlungen der Völkerwanderungszeit im nördlichen Karpatenbecken // Höhensiedlungen zwischen Antike und Mittelalter / Erganzungsbande zum Reallexikon der Germanischen Altertumskunde. 58. Berlin; New York, Simoni K. Funde aus der Völkerwanderungszeit in den Samlungen des Archäologischen Museums in Zagreb-Nalazi seoba naroda u zbirkama Arheološkog muzeja u Zagrebu // Vjesnik Arheološkog muzeja u Zagrebu. Serije 3. XXII. Zagreb, Tejral J. Naše země a Římské Podunají na počátku doby stěhování národů // Památky archeologické LXXVI. Tejral J. Zur Chronologie der frühen Völkerwanderungszeitl im mittlern Donauraum // AAuastr Tejral J. Neue Aspekte der frühvölkerwanderungszeitlichen Chronologie im Mitteldonauraum // Neue Beiträge zur Erforschung der Spätantike im mittelten Donauraum. Brno, Tejral J. Beiträge zur Chronologie des langobardischen Fundstoffes nördlich der mittleren Donauraum // Probleme der frühen Merovingerzeit im Mitteldonauraum / Spisy AÚ AV ČR. 19. Brno, Tejral J. Ein Abriss der frühmerowingerzeitlichen Entwicklung im mittleren Donauraum bis zum Anfang des 6. Jahrhunderts // Kulturwandel in Mitteleuropa. Langobarden Awaren Slawen. Bonn, Tejral J. Cultural or Ethnic Changes? Continuity and Discontinuity on the Middle Danube ca A.D. 500 // The Pontic- Danubian Realm in the Period of the Great Migraton. Paris; Beograd, Točík A. Nové nálezy z doby sťahovania národov na Juhozápadnom Slovensku // Študijné zvesti AÚ SAV. 9. Nitra, Točík A., Drenko J. Výskum v Prši na Slovensku // Archeologické rozhledy Tomka P. Über ein Haus und eine Fibel // Terra Barbarica / Monumenta Archaeologica Barbarica. Series Gemina. II. Łódź; Warszawa, Vinski Z. Archäologische Spuren ostgotischer Anwesenzeit im haitugen Bereich Jugoslawiens // Problemi seoba naroda u Karpatskoj kotline. Novi Sad, Werner J. Studien zu Grabfunden des V. Jahrhunderts aus der Slowakei und der Karpatenukraine // Slovenská Archeológia VII / 2. Werner J. Zu einer elbgermanischen Fibel des 5. Jahrhunderts aus Gaukönigshofen, Ldkr. Würzburg // Bayerische Vorgeschihtsblater Zaharia Em, Zaharia N. Les nécropolis des IV-e V-e siècles de Botoşani Dealul Cărămidăriei // Dacia XIX. Айбабин А. И. Хронология могильников Крыма позднеримского и раннесредневекового времени // МАИЭТ I. Айбабин А. И. Этническая история ранневизантийского Крыма. Симферополь, Амброз А. К. Проблемы раннесредневековой хронологии Восточной Европы. 2 // СА Амброз А. К. Боспор. Хронология раннесредневековых древностей // БСб. 1. М., Амброз А. К. Юго-Западный Крым. Могильники IV VII вв. // МАИЭТ IV.

125 Brooches of the Levice-Tokari Sub-Group 125 Борисова В. В. Могильник у высоты «Сахарная головка» // ХСб. V. Симферополь, Веймарн Е. В., Айбабин А. И. Скалистинский могильник. Киев, Виминацjум главни грод римске провинциjе Горнье Мезиjе. Београд, Гавритухин И. О. Причерноморская серия фибул типа Левице-Токари // БСб. 4. М., Гавритухин И. О. Финал традиций культур римского времени в Восточном Прикарпатье // Die spätrömische Kaiserzeit und die frühe Völkerwanderungszeit in Mittel- und Osteuropa. Łodź Гавритухин И. О. Днепровские ингумации второй половины V VI вв. // Культурные трансформации и взаимовлияния в Днепровском регионе на исходе римского времени и в раннем Средневековье. СПб., 2004а. Гавритухин И. О. Ранние формы пальчатых фибул и образец из Замятино // Острая Лука в древности. Замятинский археологический комплекс гуннского времени / РСМ. 6. М., 2004b. Гавритухин И. О. Некоторые перспективы изучения финала черняховской культуры // II Городцовские чтения / ТГИМ М., Гавритухин И. О. Днепровское лесостепное Левобережье. Ингумации, связанные с культурой оседлого населения, и отдельные находки постгуннского времени // Восточная Европа в середине I тысячелетия н. э. / РСМ. 9. М., Гавритухин И. О. Византийские подвязные фибулы с S-видной петлей для оси пружины. Находки к северу и востоку от Дуная // Археология Восточной Европы в I тысячелетии н. э. / РСМ. 13. М., 2010а. Гавритухин И. О. Поздне- и постчерняховские традиции в «Большом Киевском треугольнике» // Черняхiвська культура. Актуальнi проблеми дослiджень. Київ, 2010b. Гавритухин И. О. Фибулы типа Удине-Планис // Петербургский апокриф. Послание от Марка. СПб.; Кишинев, Гавритухин И. О., Казанский М. М. Боспор, тетракситы и Северный Кавказ во второй половине V VI вв. // АВ. 13. СПб., Гавритухин И. О., Обломский А. М. Гуннская держава и Лесостепное Подонье // Взаимодействие народов Евразии в эпоху великого переселения народов. Ижевск, Дмитриев А. В. Раннесредневековые фибулы из Дюрсо // Древности эпохи Великого переселения народов V VIII вв. М., Залеская В. И. Памятники византийского прикладного искусства IV VII веков. СПб., Зубарь В. М. Некрополь Херсонеса Таврического I IV вв. н. э. Киев, Казанский М. М. Хронология начальной фазы могильника Дюрсо // Историко-археологический альманах. 7. Армавир; М., Калитинский А. К вопросу о некоторых формах двупластинчатых фибул из России // Seminarium Kondakovianum. II. Prague, Масленников А. А. Зенонов Херсонес городок на Меотиде // Очерки археологии и истории Боспора. М., Обломский А. М. Причерноморские элементы на памятниках Верхнего Подонья середины I тыс. н. э. // Петербургский апокриф. Послание от Марка. СПб.; Кишинёв, Платар. Колекцiя пердметiв старовини родин Платонових i Тарут. Каталог. Киïв, Ушаков С. В. Варвары горной Таврики на рубеже эпох / Археологический альманах. 21. Донецк, Хайрединова Э. А. Женский костюм варваров Юго-Западного Крыма в V первой половине VI вв. // МАИЭТ IX. Хайрединова Э. А. Обувные наборы V VII вв. из Юго-Западного Крыма // МАИЭТ X. Хараламбиева А. Фибули от V в. от Североизточна България // Археология Хараламбиева А. Фибули от I VII в. в Добричкия музей // Добруджа Хараламбиева А., Атанасов Г А. Фибули от епохата на великото преселение на народите // Известия на Народния музей Варна (44). Черкун Й. Погребения эпохи Великого переселения народов возле с. Великая Бакта // Slovenská Archeológia XLII / 1.

126 126 Igor GAVRITUKHIN Игорь ГАВРИТУХИН Фибулы подгруппы «Левице-Токари» Резюме Данная статья продолжает разработки автора, начатые в 1994 г., на основе новых материалов и наблюдений. Анализируемая подгруппа включает литые фибулы длиной 6 7,5 см (известны также от 5,4 до ок. 9 см) с одним тремя выступами на треугольной или сегментовидной головной пластине и одним тремя выступами на ножке, по форме близкой ромбу, украшенные комбинациями рельефных треугольников и четырёхугольников, имеющие одну стойку для крепления оси короткой пружины и сплошной приёмник. Основная серия (рис. 1, ; 5) объединяет находки, имеющие треугольную головную пластину, украшенную четырьмя врезными треугольниками, и ромбическую ножку, украшенную четырьмя врезными ромбами; на головной пластине и ножке по 1 выступу-кнопке простой формы. Длина образцов в основном 6,0 6,6 см. Как правило, на обороте пластин по оси симметрии они имеют «ребро». Серия делится на блоки вариантов: «Левице» (с врезным овалом на дужке), с вариантами, отличающимися по форме кнопки, сегментовидной в сечении или полой дужкой и т. д.; другие (без врезного овала), с вариантами отличающимися сечением дужки, формой кнопки на ножке, пропорциями элементов, размерами и т. д. Серия показательна для фазы D2/D3. Она сформировалась на северо-западе ареала подгруппы, в герульскосвевской среде, во второй четверти V в. и распространилась в среде «окологуннских» элит в эпоху расцвета гуннской державы. Образцы паннонской серии (рис , 6. f) отличают вытянутая по вертикали головная пластина, более крупные размеры, усложненная форма выступов на ножке, в ряде случаев орнаментация пластин и наличие боковых выступов. Серия представлена тремя экземплярами, относящимися к разным вариантам, имеет компактный ареал на северо-западе Паннонии и является единственным примером развития на основе образцов основной серии в Среднем Подунавье. По стилю датируется не ранней частью фазы D2/D3 или фазой D3. Образцы дунайско-донской серии (рис , 6. a e) отличают от образцов основной серии крупная головная пластина, близкая равностороннему треугольнику, часто сложные формы выступов на ножке, боковые выступы на пластинах, более крупные размеры. Серия делится на блоки вариантов «Русенско- Алёшки» (выступ на ножке в виде трилистника) и «Сахновка» (выступ напоминает вытянутый треугольник или заостренный полуовал, в ряде случаев рельефно оформленный), вариант «Ксизово» (выступ округлый), вариант с рельефными дужкой и выступом на ножке (представлен только нелегальными находками без точной локализации), наиболее типологическими поздними экземплярами из Ботошань, у которых концы ножки утрачены. Большинство вариантов синхронны паннонской серии, но образцы из Ботошань и схожие с ними близки варианту «Херсонес», наиболее позднему в причерноморской серии. Ареал соответствует зоне группировок, оставшихся верными гуннам во время и после битвы у Недао ( гг.). Экземпляр из Керчи (рис , 6. g), вероятно, относится к особой локальной серии. Образцы причерноморской серии (рис , 7. a d) отличают сегментовидная головная пластина, более сложная орнаментация, чем у образцов основной серии, часто форма выступа на ножке, боковые выступы на пластинах, более крупные размеры. Серия делится на блоки вариантов «Токари-Сахарная головка» (выступ на ножке в виде трилистника) и «Керчь- Копачинцы» (выступ простой формы), вариант «Херсонес» (вытянутый выступ на ножке; только для него и основной серии отмечено «ребро» на обороте пластин). Для последних показательна орнаментация головной пластины выпуклыми, а не врезными треугольниками. Блок вариантов «Токари-Сахарная головка» в основном синхронизируется с паннонской серией и имеет реперный комплекс второй половины V в. Этот блок вариантов показывает формирование рассматриваемой серии в лесостепной зоне Восточной Европы и, наряду с блоком ва-

127 Brooches of the Levice-Tokari Sub-Group риантов «Керчь-Копачинцы», взаимодействие с культурами Крыма. В последние десятилетия V в. эти контакты прерываются, показателем чего является и наиболее типологически поздний локальный вариант «Херсонес». Верхнюю хронологическую границу определяет смена типов элементов убора в Крыму, связанная с событиями 530-х гг. 127 Образцы балканской серии (рис , 12; 7. е) отличает головная пластина, близкая к вытянутому по горизонтали треугольнику, длиной менее 6 см. Единичные находки на северо-востоке Болгарии являются локальными переработками образцов дунайско-донской серии и показывают угасание в регионе традиций, представленных фибулами подгруппы «Левице-Токари».

128 128 Thomas GRANE Varpelev Carsium Constantinople. Contacts across Barbaricum Introduction From the late third century AD, we see a shift in Europe of the political and cultural balance from the West to the East, a shift that is exemplified by the transformation of the Roman emperor Constantine the Great of the Greek city of Byzantion into the Imperial residence of Constantinople in AD 330. This shift is equally visible in the Germanic world, as it is in the Roman Empire. A key site for the study of this transition is the cemetery of Varpelev on Eastern Zealand traditionally dated to C2 (AD 250/60 310/20). On the one hand, the grave goods from this cemetery show a close relationship to the Germanic power centre of Himlingøje (Lund Hansen et al. 1995). This centre was founded in the middle of the second century AD and lasted until the fourth century. On the other hand, the Varpelev grave goods link to the Germanic warrior elites, as they are represented in graves from the fourth and fifth century AD. As such, the Varpelev cemetery is exemplary to the transition from the Roman to the Germanic Iron Age and at the same time a link to late Antiquity in a proto-historical part of Europe. The focus of this paper will mainly be on the chronology of the gravesite and on the supra regional perspectives of the finds, particularly those that stretch out towards the eastern part of the Roman Empire. History of the excavation The site was discovered in the summer of 1876, when a local farmer started to dig for gravel in the edge of a small natural earthen elevation about 63 x 39m large, on his lands. After hitting the remains of human skeletons and precious objects, the farmer stopped and notified the local pastor. He immediately wrote a letter to the Royal Museum for northern antiquities, later the National Museum of Denmark, listing all that had been found by the farmer. He had stumbled directly into one of the two rich graves, later known as grave alpha. For that reason, Conrad Engelhardt from the Royal Museum hastened to Varpelev, where he excavated four inhumation graves in the southern part of the elevation. At the top of the elevation in the northern part, Engelhardt identified what he believed could be the remaining 45 cm of a mound. A cross-section 1.5 m deep and 2.5 m wide, however, did not reveal anything (Archives). In June 1877, the farmer contacted the museum again. He had continued digging in the southern part and had come across two more skeletons. After this, he had continued on the top of the elevation, presumably where Engelhardt had made his cross-section. There too, the farmer had had to stop, as he had encountered some rather large stones, which he believed could be covering more finds. Engelhardt came back to excavate the rest of the elevation in July and September. Later in 1877, he was able to publish the site in a larger article on inhumation graves on Zealand and in eastern Denmark (Engelhardt 1877). The inference already made by Engelhardt in this article was that this was the burial place of a family s household. The gravesite All in all, the site contained 29 inhumation burials in 27 graves (fig. 1). The majority of the deceased were placed in a crouched position with the heads towards north to northeast. This was the normal custom for low profile graves on Zealand. Only some of these graves were equipped with any grave goods. They bodies had been placed directly on the gravel in a depth of about 1.2 m. One of the deceased (grave r) had a fractured skull, most likely from sword cuts. Four of the graves had received extra attention in the burial process regarding the position of the body and the gifts in the graves. One grave (alpha) had already been disturbed by the initial gravel digging, but two large stones and the numerous high status objects that were found in connection with this grave gave a strong indication that it had been a high profile grave. In the three remaining graves, the deceased had been laid

129 Varpelev Carsium Constantinople. Contacts across Barbaricum 129 Fig. 1. Final plan of the gravesite (National Museum Archives, Copenhagen; drawing: Julius Magnus Petersen)

130 130 Thomas GRANE to rest on their backs with the head resting on the right side and the legs slightly bent. In two of these (graves a and z), the heads were orientated towards southwest, while the third (grave y) had been placed in a more westerly direction. On Zealand, it was the custom for high status graves to be orientated more or less towards south. Grave a was discovered in a preserved depth of 2.3 m, while grave y and z were found in a depth of 2 m. Of these four, graves y and z were equipped with a few grave goods, while graves a and alpha were very richly equipped. The gravesite also contained one urn grave, which Engelhardt did not believe belonged to the Iron Age based on the type of vessel (Engelhardt 1877: 363). Unfortunately, this vessel does no longer exist. A majority of the low profile graves were situated quite closely immediately south and west of grave a on the northern half of the natural elevation (fig. 1). Engelhardt writes in his publication that this grave was situated on the highest point of the hill (Engelhardt 1877: 350). The remaining low profile graves, except one (grave i) were situated in the vicinity of grave alpha in the south end of the elevation. Situated between the two groups of graves were the high profile graves y and z. Apart from the graves, the gravesite also contained a cooking pit 2m wide and 1m deep with about 100 smaller and closely set stones (fig. 1. x). Furthermore, there were a number of small piles of stones near several of the graves. In the publication, Engelhardt stated that the gravesite had not been excavated in full, but what he meant exactly and where the gravesite would have continued is today impossible to know (Engelhardt 1877: 349). Based on the size of the gravesite and the composition of the deceased, Engelhardt believed that it had belonged to a noble family and its household (Engelhardt 1877: 350). The gender determination was based partly on anthropological and partly on archaeological remains. All in all 11 bodies could not be determined, nine were male, seven were female and two were infants. Composition and relative chronology Since Engelhardt s publication in 1877, it has been believed that the entire gravesite was contemporary and that it should be dated relatively late in relation to other graves from the Roman Iron Age. In the nineteenth century this date was established as period C2 (Lund Hansen 1987: 416). No scholars, however, have looked closely at the gravesite as a whole since the publication. Rather they have drawn out different object groups for different scholarly purposes. If we take another look at the composition of the graves, it is clear that graves y and z are situated on the central part of the elevation. They are also the only graves, which have no other graves in their immediate vicinity. Furthermore, these two graves are high profile graves, but with limited grave goods, unlike graves a and alpha. Grave y The deceased, a male, was meticulously placed on a thin layer of clay, and the body had also been packed with clay and fine sand (Engelhardt 1877: , fig. 17) (fig. 2). A number of larger stones were placed around the grave. Under the largest of the stones below the feet were found a three layered bone comb with iron rivets (Fig. 3) dated by Jørgen Ilkjær loosely to period C (Ilkjær 1993a: 309; 1993b: 489, list 45a), and a clay vessel with Fig. 2. Varpelev Sb. 8, grave y (Engelhardt 1877: 24, fig. 17) a high neck and ornamentation. Unfortunately, the clay vessel is lost today. The fact that the grave goods were found under the stone and the general placement of the stones around the grave indicates that they may have had the function of weighing down wooden boards covering the grave. Based on the central position of the grave and the lack of graves close by, this grave may have been marked in some way by a row of stones on the surface or even a small barrow. At the beginning of the summer campaign of 1877, Magnus Petersen, the museum s illustrator, made a plan and profile drawing of the elevation at a time, when only the northern part and the southeastern corner had been excavated (fig. 4). This illustration combined with the final plan using the unearthed graves zeta and a as orientation will serve to show that grave y was in fact situated at the centre of the elevation, albeit not at the highest point,

131 Varpelev Carsium Constantinople. Contacts across Barbaricum Fig. 3. Varpelev Sb. 8, grave y. Three layered bone comb with bronze rivets. Dimensions: 9.9 x 10 cm (National Museum, Copenhagen; photo: the author) in as much as we can trust the accuracy of Magnus Petersen s drawings (fig. 5). Closer scrutiny of the two plans will show some discrepancies, but regarding the placement of grave y, it is acceptable. Engelhardt writes that according to local rumours there used to be a barrow on top of the elevation. Certainly, considering Engelhardt s trial section of 1876 mentioned above, he thought it would have been in the northern part, which actually turned out to be more or less on top of grave a, which does give his thoughts some credibility. Another possibility is that it marked a founder s grave at the centre of the hill. A possible demarcation with a diameter of 12 to 15 metres would provide amble space for a small barrow. That would also mean that grave z must have been a burial secondary to y, placed near the edge of the demarcation of grave y. It should be noted though, that also grave z is without close graves, so the same argument could be made for this grave, except for the fact that it was not situated in the centre of the elevation. For that reason it is my hypothesis that grave y should be considered the earliest central grave. Grave z The deceased in grave z had been placed on a thin layer of sand (Engelhardt 1877: 364, fig. 20) (fig. 6). The gender was archaeologically determined to be a woman based on a necklace containing at least 44 plain turned amber beads. Next to the body was a bronze fibula of type Almgren VII, series 3, variation 205 with tremolo ornamentation on the loop in a zigzag pattern (fig. 7). A small clay vessel and some iron are no longer preserved. A com partment at the head end dug into the gravel 131 contained two handled clay vessels containing animal bones. The fibula, as the only datable object, places this grave in period C1b2 (Ethelberg 2009: 17 23). If we follow the logic of this hypothesis, the central grave y must pre-date grave z providing us with a rough date of that grave and the beginning of the burial place to within C1, but C1b2 at the latest provided that the two graves could be almost contemporary. As was the case with the physical requirements for the graves, i. e. the space around them, the datable material does not, however, speak against the opposite, that grave y could be younger than grave z and most probably C1b. Therefore, the hypothesis that grave y is older than grave z rests solely on the central position of grave y. Based on this, I will cautiously date the central grave to period C1b. Graves a and alpha The two remaining high profile graves were placed each in one end of the elevation, grave alpha to the south and grave a to the north. If there was a specific reason for that, it naturally eludes us. As mentioned, Engelhardt believed that there might have been a barrow on the north part of the elevation. After discovering grave a, which was the most elaborate grave by far, it is very likely that he felt confirmed in this belief. The problem is that the space around grave a, at least on the southern, western and northern sides, was densely occupied by low profile graves. For grave alpha the situation is more or less the same, although the surrounding area was less densely occupied. There is a potential problem in the fact that we have no datable material regarding the low profile graves, but they are assumed to be contemporary with the high profile graves. As we are no longer viewing the whole gravesite as contemporary, this assumption needs to be more closely defined. The low profile graves form two groups. The southern group contained six graves, of which one belonged to an infant and the rest were undefined adults. The northern group including two double graves contained 15 graves, of which one belonged to an infant, five were female and seven were male. Five were undefined. It does seem logical to infer, as did Engelhardt (1877: 350), that we are dealing with a family and its household. The two groups can be seen as linked to the two rich high profile graves. Particularly grave a is surrounded very closely by graves. One grave (b) is

132 132 Thomas GRANE Fig. 4. Plan and profile of the elevation, drawn on July the 14th (National Museum Archives, Copenhagen; drawing: Julius Magnus Petersen)

133 Varpelev Carsium Constantinople. Contacts across Barbaricum 133 Fig. 5. Combination of figures 1 and 4 showing the position of grave y in relation to graves zeta and a. Blue lines: high profile graves; blue ring: possible demarcation of central grave

134 134 Thomas GRANE Fig. 6. Varpelev Sb. 8, grave z (Engelhardt 1877: 365, fig. 20) practically on top of grave a (fig. 11). It belonged to a small but strong woman according to the anthropological material (Engelhardt 1877: 359). At the neck, Engelhardt found a piece of iron, 3.4 cm long, which he thought might have come from a neck ring. That piece of iron, which is no longer preserved, led to an idea that this woman might have been a personal slave to the person buried in grave a. This may very well have been the case, but it is perhaps more probable that the piece of iron belonged to a fibula rather than an eighteenth century style slave ring. Unfortunately, the iron fragment is not preserved. All in all, the closeness and orientation towards the two high profile graves allows for the inference that the low profile graves are more or less contemporary with them, and at least later. On the other hand, this makes it very unlikely that grave a had been marked with a barrow, as that would have left an empty space around the grave, such as was visible by the recent discovery in 2007 at Ellekilde on Zealand of a gravesite with 30 graves surrounding a founding grave leaving room for a barrow (Iversen 2011: 73 74). An interesting fact is that, if the graves were marked on the surface, all the high profile graves would have been visible from the eastern side of the elevation. Whether that is significant is, of course, impossible to know, but it could indicate that the undiscovered settlement related to the gravesite should be found in that direction. Regarding other high profile graves in the region such as Valløby, Himlingøje and Varpelev Sb. 6, situated just 1.5 km west of the Varpelev Sb. 8 gravesite, visibility appears to have mattered a great deal. Fig. 7. Varpelev Sb. 8, grave z. Bronze fibula, type Almgren VII, series 3, variation 205. Dimensions: 3.1 x 2.6 cm (National Museum, Copenhagen; photo: the author) The rich graves Now, let us turn to the contents of the two richly equipped graves. I have on previous occasions talked about these two graves as closely connected, suggesting that we may have to revise their dating (Grane 2011: 108). In the publication, Engelhardt decided to name the two graves a and alpha. This may have made sense to him, when he was preparing the manuscript, but in retrospect it was not one of his brighter moments, as it has led to some confusion concerning the origin of different objects. As mentioned, since the publication in 1877, nobody has worried much about the site as a whole, whereas the individual objects and object groups have received some attention over the years in connection with various examinations and analyses. Based on these studies, the two graves have been dated to period C2 and with them, the entire grave site. A few attempts to replace particularly grave a in the chronological context have been rejected, among other things due to methodological errors (Lund Hansen 1987: 41). However, closer examination shows that such a narrow dating is questionable.

135 Varpelev Carsium Constantinople. Contacts across Barbaricum 135 Fig. 8. Varpelev Sb. 8, grave alpha (National Museum, Copenhagen; photo: John Lee) Grave alpha We know little about the context of grave alpha as this grave was disturbed. All we know is that two large stones were involved. The gender of the deceased in grave alpha was determined archaeologically to be female. This was based on the fact that a large part of the preserved grave goods were jewellery (Engelhardt 1877: ) (fig. 8). Thus, there was a magnificent silver swastika fibula with a large amber disc in the middle and decorated with pressed gold sheet, particularly in the form of small birds. This piece has recently been placed in period C3 by Marzena Przybyła in her work on the magnificent fibulae (Przybyła 2009: 50). The gold objects consisted of two fingerings and an elaborate silver pin with a gold head. One ring is a so-called snake s head ring type Beckmann 39c. This type has been very closely connected to the theories concerning status levels of the Himlingøje dynasty. 28 examples of the Beckmann 39c ring are found, of which four were found in Norway and five in Sweden. The remaining 19 are all from Denmark. As is the case with most types of gold rings from the Roman Iron Age only a few were found in a datable context. Thus, five graves in Denmark and one in Norway contained a gold ring type Beckmann 39c. These graves have been dated to periods C1 2 (Ethelberg 2009: ; Kibernich 1995: ). The fact that the ring is rather small compared to other rings of the same type, has led Lisbeth Kibernich to Fig. 9. Gold hair pins with pelta-shaped pendants. Left: Varpelev grave alpha. Length: 6 cm. Right: Årslev (National Museum, Copenhagen; photo: John Lee)

136 136 Thomas GRANE Fig. 10. Varpelev Sb. 8, grave a (National Museum, Copenhagen; photo: John Lee) suggest that the bearer must have been very important, since a ring was specially made for her. The other ring is equipped with an onyx and decorated with various semicircular and concentric elements. It belongs to Beckmann s type 17b, which has been under discussion. Beckmann interpreted this type as a Germanic form, as it only appears in Scandinavia. Kent Andersson, on the other hand, after an examination of the craftsmanship put into the fabrication of these rings, suggests that their origin should be found in the Roman province of Pannonia in present-day Hungary. Several have since then disputed this. He suggested that they could originally have been given as gifts from Roman officials to Germanic warriors, something that is known to have happened. This is a theory that I have supported regarding this ring, as I believe that it may easily have been passed on to a woman within the family, perhaps for the simple reason that it was too small (Grane 2011: 109). The silver and gold pin is well known, and is normally compared with a pin from the Årslev grave on Fyn that is also equipped with three peltashaped pendants (fig. 9). It has been suggested that these two pins may have come from south-eastern Europe. Strangely enough, these two are rarely connected to the pin from Nyrup on Sjælland, which probably also had three pelta shaped pendants. Only Marzena Przybyła also makes this connection (Przybyła 2009: 50). The similarities suggest Zealandic relations. Possibly, the Varpelev pin may have served as inspiration for the Årslev example, but if we strip the Årslev pin of all its glamour, we may observe a close resemblance to the pins from grave A3663 from Kærup on Sjælland and grave 2223 from Møllegårdsmarken on Fyn (Christensen 2011: 32 33, Henriksen 2009: 124). When we look at these three, it is clear that there are stylistic resemblances that cannot be pure coincidence. Although the grave from Kærup is dated to the transition C1b C2, the Nyrup grave is clearly C3 (Lund Hansen 1987: 410). The Årslev grave has been somewhat disputed with datings in both C2

137 Varpelev Carsium Constantinople. Contacts across Barbaricum 137 Fig. 11. Varpelev Sb. 8, grave a, plan of the grave (National Museum Archives, Copenhagen; drawing: Julius Magnus Petersen)

138 138 Thomas GRANE to and C3 (Storgaard 1990: 49; Ethelberg 2011: ). Mogens Bo Henriksen in his analysis of the gold hoard from Brangstrup refers to both Årslev and Varpelev as parallels to some of the pelta-shaped pendants in the hoard (Henriksen 1992: 50 55). The Brangstrup hoard, which has parallels in the south eastern part of Europe, has a terminus post quem of AD due to the youngest Roman coins in the hoard. The conclusion is that the pins from both Varpelev and Årslev may very well be highly elaborate works of local craftsmen, but with elements from the Southeast in the form of pelta shaped pendants. Grave a Let us now turn to grave a, which has by far the most spectacular grave ensemble (Engelhardt 1877: ) (fig. 10). The deceased was laid to rest on his back at least 3 metres below ground level in a stone coffin so to speak, consisting of 16 large natural stones, creating a chamber of 5.3 x 1.80 m (fig. 11). Most of the skeleton had dissolved, but based on an examination of the thigh and arm bones, the deceased was determined as a heavy limbed, strongly developed male (Engelhardt 1877: 352, note 1). The body was situated in the middle with room for grave goods both above his head and beneath his feet. The space between the individual stones surrounding the grave had been closed with clay, forming an actual chamber, a sarcophagus, so to speak. The foot end of the grave had been covered with a layer of small stones. Furthest north on this layer stood a Roman bronze basin type Eggers 105 with parallels in Årslev and Søndre Kjørstad that had been placed on top of a wooden bowl containing bones of a pig. Inside the basin was a three layered bone comb datable to C2 (Ilkjær 1993b: 490, list 45a). Everything had been covered with some kind of textile. Next to the basin was a large wooden bucket with bronze fittings. Both inside and outside of the bucket, 42 gaming pieces of bone were found; some of these also under the flat stone together with some animal bones. Between the feet of the deceased were found the remains of four bear s claws. In the southern end of the grave, a large flat stone had crushed most of at least six glass vessels and one ceramic vessel, as well as a horn with silver fittings. Of the six glass vessels, a number of fragments belong to one possibly two ordinary facetted glass vessels type Eggers 230. The remaining four vessels are all unique. The green facetted glass has a height of 18 cm. It has five rows of ovals and a double concave rim. There is today, some uncertainty concerning the origin of the facetted glasses, as some are believed to be of a Gothic Black Sea origin, while others are thought to be Roman. According to Ulla Lund Hansen, the very high quality of craftsmanship applied to this piece indicates an origin in a workshop in the region of Constantinople. A major problem has been finding parallels. The only vessel with resembling elements comes from the incredibly rich second grave from Apahida near Cluj in Romania (Goldhelm 1994: 249, no ). This grave is dated to the middle or second half of the fifth century, but the glass is likely to have been an antique, as it has been repaired with gold foil. The glass is 16 cm high and has three rows of somewhat larger ovals with a small spacing in between. Particularly the double concave rim is alike on the two glasses. An interesting parallel to the Apahida piece comes from the unpublished necropolis of Carsium, a Roman city on the Danube near the Black Sea. Like the one from Varpelev, it is 18 cm high (Moderán 2008: 149, cat. no. 663). The facetted area is very similar to the Apahida piece, although the spacing is a little different. Below the rim, there is a zone with a Greek inscription, ПΙΕ ΖΗCΑΙC ΕΝ ΑΓΑΘΟΙC meaning Drink and You shall live among the Good. This is clearly a vessel originating in the Roman empire. The combined evidence from these three glasses supports the assumption made by Ulla Lund Hansen regarding an origin of the Varpelev glass in the region of Constantinople or the eastern Roman provinces. In late Antiquity, this, and other similar inscriptions are used in gold decorations of the bottom of glass vessels especially by the Christian community, who, for instance, used these gold decorated bottoms to mark and personalize cells in the Catacombs. The highlight of the grave goods is without a doubt a kantharos made of blue glass blown into an openwork silver frame decorated with wine leaves and the Greek inscription ευτυχωc meaning For your happiness (fig. 12). According to Susanna Künzl, this work is probably of a Syrian or at least an east Roman origin (Künzl 2000a: 76; 2000b: 614). There are a couple of parallels to the technique, although not to the motif. The closest parallel is dark red and comes from Georgia, now at the Ermitage in Saint Petersburg (Birjukowa 1986: no. 6).

139 Varpelev Carsium Constantinople. Contacts across Barbaricum 139 Fig. 12. Varpelev Sb. 8, grave a. Detail of Greek openwork inscription ευτυχωc meaning For your happiness on silver/glass kantharos (National Museum, Copenhagen; photo: the author) Künzl also established that the Varpelev kantharos had been repaired. Thus the top plate that served as a thumb handle did not belong originally. Parallels to the top plate have been found in Georgia (Künzl 2000a: 76). Furthermore, the index finger rings had been substituted by the middle finger supports for some reason. Apparently, Künzl did not realize that the two substituted rings were in fact part of the grave goods. This indicates two things. First of all, that this vessel had been used regularly; secondly, that the vessel was sufficiently important for the two rings to have survived to be put in the grave. What then, could be the reason for the alteration? Well, a look at the measures of the rings, show an internal diameter of mm. The altered handles of the kantharos have an internal diameter of mm, corresponding with the two gold fingerings that were also found in grave a. The most enigmatic of the glass vessels, has for long been considered a pipette of sorts as it is open in both ends. We actually know of Roman glass pipettes, but they are much more delicate, why such a use is highly unlikely. The closest we can come to this shape, are unguent containers belonging to the fourth and fifth century, but they are always closed in one end, and in some cases they have even been closed in the other end after the filling of perfumed oil. This type of vessel is solely found in mortuary contexts, and often with a glass vessel with an inscription. This has led Hillary Cool to suggest that they belong to some ritual to do with religions believing in an afterlife such as the Bacchus cult or Christianity (Cool 2002). She has suggested that the Varpelev example must have been specially made, as she has seen nothing like it. That means that the container would have had to be sealed in both ends. It is not entirely obvious that it is this type of container, but the combination with a vessel with an inscription at least allows for such a scenario. On his right hand s thumb and ring finger, he had two rather common gold rings. At his right shoulder there was a small gold pin. A snake s head arm ring type C was located near the neck. This type is considered by Per Ethelberg to originate in Central Sweden due to the resemblance

140 140 Thomas GRANE with snake s head fingerings type 40, which are mainly distributed in Sweden and Norway, while only a few are found in Denmark (Ethelberg 2011: 115). These rings, however, show great variety in form. Furthermore, only a few are actually found in a context. Kent Andersson, on the other hand, believes that this type is produced locally based on his examination of workshops in Scandinavia, something that he underlines using precisely the Varpelev ring (Andersson 1995). On the right side of the head there was an aureus from Emperor Probus, who reigned from AD 276 to 282, equipped with a rather crude loop. I have on previous occasions used among other things this coin to present a scenario, which mirrored the presence of aurei from the Gallic emperors in graves from the Haßleben-Leuna region, that is, that we may see an attempt to reinstate conditions regarding relations to the Romans as they were before the crisis in the 260s 270s (Grane 2007, 276). This coin can only provide us with a terminus post quem of the early 280s. At the waist, there was a silver belt buckle with a strap end fitting and a couple of smaller fittings. The loop is decorated with a cell pattern and the tongue with an animal s head (fig. 13). The rivets are also decorated. The buckle is a late Roman military belt type. Late Roman military belts are found in most of the Roman provinces, especially Gaul and Pannonia and belong to the fourth and fifth century. The Varpelev belt belongs in Markus Sommer s group A1a and Erwin Keller s A B C buckles. This type is dated to AD 290 to 400 for the Danube region by Sommer and more narrowly to 330 to 360 by Keller (Keller 1971: 58; Sommer Fig. 13. Varpelev Sb. 8, grave a. Silver belt buckle and strap end. Dimensions: loop 4.5 x 3 cm; plate 4.5 x 2.8 cm; pin 2.9 x 0.5 x 0.4 cm; strap-end 3.4 x 2.1 cm. (National Museum, Copenhagen; drawing: the author)

141 Varpelev Carsium Constantinople. Contacts across Barbaricum 1984: 74 75). The strap end is decorated similarly, but the cells appear less elaborate, indicating that it may not have been made by the same or at the same time as the buckle. As the cell ornamentation is practically unknown on Roman, or other belts for that matter, several scholars have considered the belt a Germanic copy rather than a Roman belt. Michel Kazanski considered this belt in an article in the conference publication of Inter Ambo Maria from 2011, where he demonstrates a relation of the Varpelev belt and other artefacts with cell ornamentation to a Roman tradition, arguing that the belt had been produced in a Roman workshop and most likely under orders by Barbarian chiefs (Kazanski 2011). As is the case with the other unique objects in this grave, it is extremely difficult to find good parallels. Some examples have been suggested to me by colleagues coming from as different places as Krefeld-Gellep on the Rhine, Pruszcz Gdánski in Poland, Zhitomir in Russia, Augsburg in Germany and Spanţov and Hârşova in Romania. On a general level, all these buckles, and in fact many others on both sides of the Roman frontier have a slight likeness to the Varpelev example. But most of them only have a superficial resemblance, when it comes to possible inspiration and origin. In fact only the belt buckles from Zhitomir (Kazanski 2011: 98, fig. 5. 5) and Spanţov (Mitrea, Preda 1966: 20, 207, fig ) (fig. 14) are good parallels to the main buckle. The loops on both are made with cell ornamentation like the one from Varpelev, although they have respectively more and fewer lines of cells. It is therefore, quite clear that the search for parallels should focus on details rather than entire buckles. For the best detail parallels we have to look closer at a number of belt buckles and fittings from the unpublished Roman necropolis of Carsium, present day Hârşova (Goldhelm 1994: 225, no. 94; Moderán 2008: 149, cat. nos , ). These fittings have been dated to AD 320 based on coins found in a purse with the deceased (Kazanski 2011: 95, note 3). If we compare the different elements, there are a lot of resemblances. The loops of the main buckles have the same dimensions, although the Carsium example is ornamented with animal s heads, probably wild boars. The elements of the tongue are identical with a triangular cross section. At the hinge it has a rectangular or square part to block the plate, and the point is shaped as an animal s head. 141 Some of these features are the same on some of the smaller buckles. The shape of these is much similar to the small Varpelev buckle, although it has only one rivet, something that Kazanski (2011, 96) points out as well. The hinge ornamentation on one of the small buckles consists of two lines parted by a roof-shaped zone, something that is also hinted at on three of the smaller fittings, which, along with an undecorated piece, have the exact same shape as the small fitting from Varpelev (Kazanski 2011: 94, fig ). And last but not least, the strapend is equipped with three clover-shaped rivets that could be construed as inspirational to the rivets from Varpelev. In general, no late Roman belts with decorated rivets have come to my attention, except this. The fact that the parts are made of silver, only enhances the impression. I have, therefore, reached the same conclusion as Kazanski that the Varpelev belt was made to order by craftsmen inside the Roman Empire, most likely in the eastern Danube region (Grane 2013: 191). The dating of the gravesite I have here attempted to present the gravesite of Varpelev Sb. 8 in a way that makes it clear that it is a much more complex site than has been presented before. It is possible to see a development in the gravesite from one central founding grave (y) in the centre of a natural elevation. The care that was shown in the burial practice indicates that this person was important to the local community, a part of the nobility. It does not appear Fig. 14. Spanţov, grave 4. Bronze belt buckle. Dimensions: 3.3 x 1.9 cm (National Muzeum of Romanian History, inv. no ; photo: the author)

142 142 Thomas GRANE to have been a rich community or nobility judging by the grave goods in the early graves. The male deceased in grave y was probably followed by the female in grave z not to long after. Grave z belongs in the period C1b2, that is, around the middle of the third century. When the next high profile grave was being prepared, the situation had changed. That much we can assume from the wealth of both graves alpha and a. Several objects from both of these graves point towards a date in the fourth century, and most likely in period C3 rather than the end of C2. On the one hand, recent work on the swastika fibulae by Marzena Przybyła supports a C3 date for grave alpha. Also the pelta shaped pendants point towards the fourth century. Lisbeth Kibernich has argued that the snake s-head fingering from this grave was exceptional compaired to other Beckmann 39c rings because it was custom-made. Kazanski already demonstrated how the belt parts in grave a belong in C3, but the rest of the precious objects cannot help us with the dating, although the aureus from AD also indicates a later date according to Helle Horsnæs (2010: 60). Per Ethelberg argues strongly for a C2 date for both graves. Some important reasons given by Ethelberg are, first of all, that the snake s head rings would then last longer than the rich graves at Himlingøje. As this type of ring is strongly connected with the sphere of interest of the Himlingøje dynasty, the ring could not exist after the demise of that dynasty. Furthermore, he argues that despite Przybyła s research, four-armed swastika fibulae such as the example from grave alpha are only dated to C2. Thirdly, he argues that there was a discontinuity on Zealand between C2 and C3, that is, rich graves from C3 are never found at the same sites as rich graves from C2 (Ethelberg 2011: ). It is quite clear that based on these points it is impossible to date the Varpelev graves to C3 instead of C2. Lastly, Ethelberg believes that the gold arm ring from grave a was produced in middle Sweden and therefore not part of the group of Himlingøje insignia that are normally known as the snake s head rings. Therefore, he suggests the possibility of a power coming down from Sweden to take hold at Varpelev, because this would explain, why Himlingøje ended abruptly during the course of C2, why a Swedish arm ring was found in Varpelev and why there is a discontinuity in the rich grave environments between C2 and C3 (Ethelberg 2011: 115). This development is construed differently by Ulla Lund Hansen, who saw Varpelev as the natural continuation of Himlingøje (Lund Hansen et al. 1995: 392, 430). Furthermore, the so-called demise of the high profile graves may have something to do with a change in grave customs and rituals. In fact, it is worth noting that in the Eastern Zealand/ Stevns area it is possible to observe continuity in the rich finds, despite the changed depositional practices in the transition from the Late Roman to the Early Germanic Iron Age (Lund Hansen 2001: 227). This is not supporting a sudden demise of Himlingøje. Quite clearly, the situation in the end of the third and beginning of the fourth century is nothing if not complicated. In fact, the transition from C2 to C3 may stand considerably clearer in our archaeological chronology systems, than it did for members of the Iron Age societies of that time. Ethelberg s arguments primarily fit into artificially created boxes, into which we need to put the different items in order to sort them out. The problems arise, like in the case of the rich Varpelev graves, when the finds defy the boxes. As mentioned above, Kent Andersson has argued that the snake s head arm rings were made by different craftsmen in different regions using the ring from Varpelev as his prime example (Andersson 1995). Ethelberg compares the arm ring type C with the fingering type 40, which is mainly found in middle Sweden (Ethelberg 2000). The problem is that only six examples are found and none of these are in a context. Therefore, Andersson s conclusion is preferable. Ethelberg s other arguments are expressing a need to keep things within the given rules, but the statistical material is far too little to claim that a swastika fibula has to have a fifth arm in C3. While we may be content at the time with suggesting that the fibula is drawing grave alpha towards the C2/C3 transition, the Late Roman military belt is placing grave a in the first half of C3. Conclusion The overall conclusion to the question of chronology must be that what we see in Varpelev is a gravesite, which is initiated probably in the middle of C1b by the burial of a member of the local nobility, someone, whose status perhaps does not allow for valuable gifts like his contemporaries at nearby Himlingøje, but someone who is still worthy of a burrow or demarcation. For several generations the head of the family is given the proper ceremony. At the same time, during C2, status is slowly acquired in the region with connections to an over re-

143 Varpelev Carsium Constantinople. Contacts across Barbaricum gional power. It all ends sometime in the end of the first half of C3 corresponding to the Constantinian era with the burial of a man, who had obtained the highest rank through his commitment in Roman- Germanic affairs. A large part of the grave goods in both rich graves underline an increasing contact to the eastern parts of the Empire. Whereas the rich graves of the Himlingøje gravesite revealed unique, but generally anonymous objects befitting certain demands or rituals, which involved feasting and involving drinking vessels, containers and ladles, it is quite clear that the Roman objects of particularly grave a represented a different kind of contact. While the earlier graves represented a long distance relationship, grave a is evidence of a close encounter. As mentioned above, there is a large number of unique objects that link to the eastern parts of the Roman empire; a military belt with links to the Danube 143 provinces and glass vessels pointing in the same direction; a very used and repaired drinking vessel with a Greek inscription; gold jewellery that reflect the highest power in the society. All this suggests that this person unlike his predecessors had to be outgoing in order to preserve his position in the society. That brought him particularly to the eastern provinces of the Roman empire, from where he brought back highly personalized items that required a personal knowledge of Roman customs. This may have been in corporation with or in spite of regional powers, perhaps to regain a lost position. Such an action would have required an active foreign policy, as it would have been clear to the Zealandic élite that their former position was gone for good with the crisis and the coming of the tetrarchy, when a new policy regarding the Germanic tribes was instigated rendering former alliances useless. Bibliography Andersson K. Romertida Guldsmida. III. Stockholm, Archives of the National Museum of Denmark. Birjukowa N. Angewandte Kunst in der Eremitage. Leningrad, Christensen L. M. Kærup Nord en gravplads fra yngre romersk jernalder ved Ringsted // Aarbøger for nordisk Oldkyndighed og Historie [no place], Cool H. E. M. Bottles for Bacchus? // Artefacts and Archaeology. Aspects of the Celtic and Roman World. Cardiff, Engelhardt C. Skeletgrave paa Sjæland og i det østlige Danmark // Aarbøger for nordisk Oldkyndighed og Historie. Kjøbenhavn, Ethelberg P. Zealand and the Roman Empire // The Iron Age on Zealand. Status and Perspectives / Nordiske Fortidsminder Serie C. 8. Copenhagen, Ethelberg P. Die Fibeln // Wealth and Prestige. An Analysis of Rich Graves from Late Roman Iron Age on Eastern Zealand, Denmark. Studier I Astronomi Nyere Tid Arkæologi. II. Taastrup, Ethelberg P. Skovgårde. Ein Bestattungsplatz mit reichen Frauengräbern des 3. Jhs. Auf Seeland / Nordiske Fortidsminder Series B. 19. Copenhagen, Grane T. A Late Roman Military Belt from Zealand? // Radovi Proceedings Akten. XVII Roman Military Equipment Conference, Zagreb Zagreb, Grane T. Zealand and the Roman Empire // The Iron Age on Zealand. Status and Perspectives / Nordiske Fortidsminder Serie C. 8. Copenhagen, Grane T. The Roman Empire and southern Scandinavia a Northern Connection! / PhD dissertation. Copenhagen, Goldhelm, Schwert und Silberschätze. Reichtümer aus 6000 Jarhen rumänischer Vergangenheit. Frankfurt-am- Main, Henriksen M. B. Brudager Mark en romertidsgravplads nær Gudme på Sydøstfyn / Jyske Jernaldergrave Fynske Studier. 22. Odense, Henriksen M. B. Brangstrupfundet. En guldskat fra slutningen af romersk jernalder // Fynske Minder. Odense, Horsnæs H. Crossing Boundaries. An analysis of Roman coins in Danish contexts. Vol. 1 Finds from Sealand, Funen and Jutland / Publications of the National Museum. Studies in Archaeology and History. 18. Copenhagen, 2010.

144 144 Thomas GRANE Ilkjær J. Illerup Ådal. 3, Textband. Die Gürtel. Bestandteile und Zubehör. Aarhus, 1993a. Ilkjær J. Illerup Ådal. 4, Tafelband. Die Gürtel. Bestandteile und Zubehör. Aarhus, 1993b. Iversen R. Ellekilde en gravplads fra yngre romersk jernalder med fyrstegrav og cirkusbægre // Aarbøger for nordisk Oldkyndighed og Historie (2009) Kazanski M. Kishpek, Ekazhevo and Varpelev: on the problem of Pontic-Scandinavian relations in the Late Roman period // Inter Ambo Maria. Contacts between Scandinavia and the Crimea in the Roman Period. Kristiansand; Simferopol, Keller E. Die Spätrömische Grabfunde in Südbayern. München, Kibernich L. Nordische Tierkopffingerringe und ihre Verbreitung // Saalburg Jahrbuch Künzl S. Quellen zur Typologie des römischen Tafelsilbers // Sborník Národního Muzea v Praze. Praha, Lund Hansen U. The Iron Age on Zealand. The Early Iron Age-graves, social structure, new perspectives // The Iron Age on Zealand. Status and Perspectives / Nordiske Fortidsminder Serie C. 8. Copenhagen, Lund Hansen U. Römischer Import im Norden / Nordiske Fortidsminder Serie B. 10. Copenhagen, Lund Hansen U. et al. Himlingøje Seeland Europa. Ein Gräberfeld der jüngeren römischen Kaiserzeit auf Seeland, seine Bedeutung und internationalen Beziehungen / Nordiske Fortidsminder Serie B. 13. Copenhagen, Mitrea B., Preda C. Necropole din secolul al IV lea E. N. în Muntenia / Biblioteca de Arheologie. X. Bucharest, Modéran Y. Die kontrollierte Einwanderung von Barbarengruppen in das Römische Reich // Rome und die Barbaren. Europa zur Zeit der Völkerwanderung. Bonn, Przybyła M. Die Hakenkreuzfibel aus dem Grab 4 aus Engbjerg auf dem Hintergrund anderer pressblechverzierter Hakenkreuzfibeln // Wealth and Prestige. An Analysis of Rich Graves from Late Roman Iron Age on Eastern Zealand, Denmark / Studier i Astronomi Nyere Tid Arkæologi. II. Taastrup, Sommer M. Die Gürtel und Gürtelbesläge des 4. und 5. Jarhunderts im römischen Reich. Bonn, Storgaard B. Årslev-fundet et fynsk gravfund fra slutningen af yngre romersk jernalder // Aarbøger for nordisk Oldkyndighed og Historie Томас ГРАНЕ Варпелев Карс Константинополь. Контакты через Барбарикум Резюме Начиная с конца III в. в Европе наблюдается сдвиг политического и культурного баланса с Запада на Восток, сдвиг, примером которого является превращение римским императором Константином I греческого города Византия в императорскую резиденцию Константинополь в 330 г. Точно также, как в Римской империи, данный сдвиг наблюдается и в германском мире. Ключевым памятником для изучения этих изменений является могильник Варпелев в восточной Зеландии, традиционно датируемый фазой С2 (250/60 310/20 гг.). С одной стороны, инвентарь двух погребений этого могильника указывает на близкую связь с германским центром власти в Химлингёе, основанным в середине II в. С другой стороны, погребальный инвентарь обозначает связи с германской военной знатью, поскольку он представлен в погребениях IV V вв. Могильник Варпелев сам по себе является примером перехода от римского к германскому времени, но в то же время проявляет и связь с поздней античностью в доисторической части Европы. Могильник Варпелев расположен на полуострове Стевнс на восточном побережье Зеландии. Его открыл в 1876 г. некий крестьянин, копавший землю в поисках гравия. Раскопки велись в гг., а результаты их были незамедлительно опубликованы в журнале «Aarbøger for nordisk Oldkyndighed og Historie» (Engelhardt 1877). Благодаря близости

145 Varpelev Carsium Constantinople. Contacts across Barbaricum этого памятника к богатым некрополям знати в Химлингёе и Валлёбю, от которых его отделяют всего несколько километров, его считают частью восточнозеландского центра власти позднеримского времени. Могильнику Химленгёе посвящено подробное исследование У. Лунн- Хансен, изданное в 1995 г. (Lund Hansen el al. 1995). После публикации могильника Варпелев в 1877 г. никого особенно не интересовал памятник в целом, тогда как отдельные предметы и группы предметов привлекали определённое внимание на протяжении многих лет, становясь объектом различных анализов и исследований. Основываясь на этих работах, могилы, а по ним и погребальный памятник в целом, датируют периодом С2. Тем не менее, новые исследования показывают, что период существования данного могильника вполне мог растянуться ещё на несколько поколений. Могильник состоял из 28 могил по обряду трупоположения, причём четыре погребения выполнены по более сложному обряду. Умерших укладывали на спине, головой на юг, что, как мы знаем, обычно для погребений знати римского времени. Прочих погребённых, насколько можно судить, укладывали в скорченном положении на боку, головой на север. В двух из четырёх особых могил находился богатый инвентарь, включавший римские импорты, а также изделия из золота и серебра (погребения а и «альфа» рис. 1 2), тогда как в двух других найдены немногочисленные обычные вещи, например, костяной гребень и бронзовая фибула. Вероятнее всего, последние являются самыми ранними погребениями на памятнике, тогда как богатые могилы относятся к следующему поколению или поколениям. Самым поздним является погребение а (рис. 1). Это самое богатое из всех погребений знати этого времени в Скандинавии. Находки включают римские сосуды шесть стеклянных и один бронзовый, наручный браслет, два перстня и булавку из золота, денарий императора Проба ( гг.), пояс с серебряными украшениями позднеримского типа, сигнальный рог с серебряными оковками, деревянное ведро с бронзовыми оковками, костяные гребень и игральные фишки, четыре медвежьих 145 когтя и керамический кувшин. В статье основное внимание уделено римским вещам. Четыре из шести сосудов уникальны, один принадлежит к довольно обычному типу 230 по Х.-Ю. Эггерсу, а ещё один утрачен. Наиболее знаменит канфар из синего стекла, изготовленный в технике выдувания, в ажурной серебряной оправе. Некоторые из этих сосудов указывают на восточную часть Римской империи. Также интересны серебряная пряжка и накладки на пояс. Тип пряжки, очевидно, римский, но её детали указывают на германское влияние. Пояс находит не так много близких параллелей фактически, есть только две действительно похожие на него находки. Первая из них небольшая пряжка из могильника Спантов в Румынии, который относится к черняховской культуре (культуре Сынтана-де- Муреш). Вторая группа пряжек и накладок из одной из могил некрополя римского городка Карс, ныне Харсова, также в Румынии. На близость ремня из Карса к нашей находке указывает сходство индивидуальных деталей, а не пряжек в целом, которые на самом деле не так и похожи. Могила в Карсе условно датируется 320 г. Подобная датировка погребения а поставит его в рамки перехода от периода С2 к С3. Из анализа могильника Варпелев и его погребального инвентаря в сравнении с содержимым ранних могил знати в Химлингёе и на других выдающихся памятниках Севера становится ясно, что вещи из могил в Варпелеве указывают на более тесные контакты с римлянами, чем ранние погребения. Если обратиться к истории Римской империи от кризиса середины III в. до периода Константина I, то становится очевидно, что менявшаяся политическая ситуация в западных провинциях должна была оказать существенное влияние на все германские племена, взаимодействовавшие с римлянами. Любые модели контактов, которые могли происходить до кризиса, должны были пересматриваться в новых условиях. Что же касается центра власти в восточной Зеландии, желание поддерживать богатый уровень жизни, как раньше, потребовало бы от знатных семейств гораздо более активных действий, отражением которых могли быть погребения а и «альфа» в могильнике Варпелев.

146 146 Eszter ISTVÁNOVITS, Valéria KULCSÁR Sarmatians or Germans: influence or ethnic presence? Old problem new hypotheses One of the long debated questions in the research of the Barbaricum of the Carpathian Basin is the appearance and ethnic background of a cultural group in the Upper Tisza region. According to both the written sources and archaeological material Vandalic groups settled at this area next to the Sarmatians during or after the Marcomannic Wars (fig. 1). The situation in the Upper Tisza region and, at the same time, in the Barbaricum of the Carpathian Basin had undergone a major change in the second half of the second century. At the start of the war, From the borders of Illyricum even into Gaul, all the nations banded together against us the Marcomanni, Varistae, Hermunduri and Quadi, the Suebians, Sarmatians, Lacringes and Buri, these and certain others together with the Victuali, namely, Osi, Bessi, Cobotes, Roxolani, Bastarnae, Alani, Peucini, and finally, the Costoboci (Historia Augusta Marc. Ant. 12). 1 In AD 171 or 172, the governor of Tres Daciae concluded an alliance with the Hasding Vandals: The Astingi, led by their chieftains Raüs and Raptus, came into Dacia with their entire households, hoping to secure both money and land in return for their alliance. But failing of their purpose, they left their wives and children under the protection of Clemens, until they should acquire the land of the Costoboci by their arms... (Cassius Dio Hist ). In exchange for receiving land, the eastern Germanic Hasdingi defeated the Costoboci of the free Dacians probably somewhere along the Lower Danube who swept into Greece through Dobrogea from their territories in the north-eastern foreland of the Carpathians. 2 Something happened upon their return because the Vandals again attacked the province. This time, the Romans sought out the Lacringi, a people related to the Vandals (or a tribe of them), as their allies:...upon conquering that people, they proceeded to injure Dacia no less than before. The Lacringi, fearing that Clemens in his dread of them might lead these newcomers into the land which they themselves were inhabiting, attacked them while off their guard and won a decisive victory. As a result, the Astingi committed no further acts of hostility against the Romans, but in response to urgent supplications addressed to Marcus they received from him both money and the privilege of asking for land in case they should inflict some injury upon those who were then fighting against him (Cassius Dio Hist ). The Romans thus succeeded in pacifying the two Vandal tribes in the foreland of Dacia. The earliest finds of the Przeworsk culture in the Carpathian Ukraine warrior cremation graves can be dated to the first century (stage B1), indicating that contacts between the Germanic peoples and the region s probably Dacian population had been established at a relatively early date (Kobály 1998). However, the overwhelming majority of the Przeworsk assemblages, e. g. the grave from Kékcse (Szabolcs-Szatmár-Bereg county) (fig. 2) from the Upper Tisza region can be dated to time of the Marcomannic Wars (stage B2 C1), i. e. to the very period when the written sources report the arrival of various Vandalic or related tribes, name- 1 The Victovali and the Lacringi first appear in the sources at this time (for a discussion of this passage and its problem, together with the earlier literature, see: Alföldi 1942: note 216). 2 For the incursion, see: Alföldi 1942: 194.

147 Sarmatians or Germans: Influence or Ethnic Presence? Old Problem New Hypotheses 147 Fig. 1 Ethnic situation in the Carpathian Basin, especially in the Upper Tisza region. a Sarmatians; b Przeworsk culture

148 148 Eszter ISTVÁNOVITS, Valéria KULCSÁR Fig. 2 Typical Przeworsk grave finds from the cremation burial at Kékcse ly the Victovali, the Hasdingi and the Lacringi, to province Dacia s north-western foreland. 3 Appearance of Germanic groups in the region can perhaps be traced in the archaeological material of the early Roman Age from the Szatmár-Bereg Plain (border area of present day Hungary, Romania, and Ukraine) brought to light as part of intensive research during recent years. The assemblages un- 3 The relevant passages of the written sources and its problems, with the earlier literature, are discussed by: Alföldi 1942: note 216; Schmidt 1942: 6 8; Bóna 1986: 61 63; Godłowski 1993; Istvánovits, Kulcsár 1994; Kobály 1998.

149 Sarmatians or Germans: Influence or Ethnic Presence? Old Problem New Hypotheses 149 Fig. 3. Hortobágy-Poroshát barrow II / 10 (Zoltai 1911: Taf. III. 10, 12 14, 18 37) covered during the excavations at Csengersima / Petea (Hungarian-Romanian border: Szabolcs- Szatmár-Bereg county in Hungary and Satu Mare county in Romania) (Gindele, Istvánovits 2009), Beregsurány (Szabolcs-Szatmár-Bereg County) (Istvánovits 2004) and recently on settlements in County Satu Mare in Romania (Németi, Gindele 1997) reflect a strong mix of Dacian and Germanic artefact types. Judging from the historical events reconstructed on the basis of the literary sources it is hardly surprising that finds such as the typical male burials containing weapons of the Przeworsk culture and culturally attributed to the Vandals, especially the Victovals, appeared in the region in the late-second century (stage B2/C1). In Hungary, the finds from the Upper Tisza region were first reviewed by István Bóna, who identified these Germanic assemblages as the heritage of the Victovals (Bóna 1986: 61 65). The chronology and typology of the type finds (shields, spurs, swords, etc.) brought to light from the cremation burials of the Przeworsk culture have been reliably established by Polish research. Kazimierz Godłowski s excellent overview of the archaeological evidence and the literary sources, also affecting the finds from Hungary is especially important for us (Godłowski 1993: 69 78).

150 150 Eszter ISTVÁNOVITS, Valéria KULCSÁR A new find horizon can thus be distinguished in the Upper Tisza region in the late-second century, made up in part of the second wave of the Przeworsk culture (sites Kékcse, Tiszakanyár, Vásárosnamény-Hajnal Street all in Szabolcs- Szatmár-Bereg county), and in part of the heritage of the Sarmatians who, according to the current evidence, occupied in mass this region at this time. 4 The Sarmatians advanced north-eastward in the Upper Tisza region and toward the river Ier in the south, reaching the Carei (Satu Mare county, Romania) area (Németi, Gindele 1997). 5 It seems likely that the Carpathian Barrows culture, identified with the free Dacians, too made its appearance in the Carpathian Ukraine during this period (Русанова 1993: 172, карта 30). However, we are not able to follow the further history of the Vandals in this region, because the horizon of cremation burials soon disappears and there is no archaeological evidence for the Przeworsk population of the northern part of the Hungarian Plain. Traditionally Hungarian and international research tried to identify its archaeological traces from the third century with the so-called Geszteréd-Herpály-Hortobágy group. The common features of the burials from these sites are that these are Sarmatian barrow graves dated most probably after the mid-third century AD with some finds of Germanic (Przeworsk) character. Among the latter, the elements of weaponry are usually emphasised. According to a general view accepted long ago, Sarmatian gravegoods do not include shields. This applies also to the earliest (post-marcomannic Wars) Sarmatian burials of the Upper Tisza region the inhabitants of which were, in all probability, neighbours with Przeworsk people living on the other side of the Iranian-Germanic border (fig. 1). Mihály Párducz, the first to interpret the Hortobágy- Poroshát (Hajdú-Bihar county) cemetery has considered it to be Sarmatian (Roxolanian) (Párducz 1950: ). István Bóna thought this group to be of Sarmatian (Alanic) origin with Germanic features (Bóna 1986: 63 66). Appearance of Germanic weapons in the burials made several researchers neglect the fact that most of the features of the burial rite referred to Sarmatians. It is important to emphasise that the number of German objects is extremely low: for example, in the 220 graves of the Hortobágy-Poroshát cemetery a single shield boss, a pair of shears and some further objects of uncertain origin can be listed as ones belonging to Germanic material culture (fig. 3). However, Mihály Nagy (1994: , ) and Péter Prohászka (2008: 50 55) in his recent work simply attributed the sites in question (the so-called Geszteréd-Herpály-Hortobágy group) to Vandals. According to Marek Olędzki (1999: 129) this group indicates a Sarmatian-Germanic culture. The main idea of these researchers is based on István Bóna s hypotheses about the gradual assimilation of Vandals of the Upper Tisza region, who, according to him, changed their burial rite (incineration to inhumation), way of decorating the costume etc. This would explain where did the archaeological traces of Vandals disappeared. However, there is no explicit evidence for the actual presence of Germans at the northern part of the Sarmatian Barbaricum in the second half or end of the third century. At the same time, there is no doubt that the group in question represents at least partly a new wave, new archaeological horizon of Iranian population (either Roxolans, or Alans or any other Sarmatian tribe or group). An interesting observation is that Przeworsk elements appeared in the Crimea in the same period. It cannot be a coincidence that one of the typical Germanic finds coming from Neyzats, vault 306 is a shield boss, belonging to the same type (Ikljær 3b, Jahn 7a see Храпунов 2011: 8) as the one from Hortobágy-Poroshát, barrow II/10. Similar umbos from the same region of Sarmatian dominance are known also from Nyíregyháza-Árpád Street, a grave with typical Sarmatian burial rite (south-north orientation, inhumation), and from stray finds of Geszteréd and Tiszalök-Rázompuszta (Szabolcs-Szatmár-Bereg county) (both from the territory of Sarmatian dominance) (Istvánovits, Kulcsár 1992: 55, 32, 50, 63). The most wellknown piece of this type is the one from Herpály (Berettyóújfalu, Hajdú-Bihar county). Judging from the burial rite (barrow, horse skeleton) and the context (neighbouring barrows with typical Sarmatian grave-goods, e. g. beads) it should be considered a Sarmatian burial with a German prestige object (Istvánovits, Kulcsár 1994: 416). 4 There are some new finds coming from large scale excavations in the vicinity of Nyíregyháza that may modify this picture, but by the time being they are unpublished, so here we use the traditional point of view. 5 The so-called Sarmatian buckle from Carei/Nagykároly (Párducz 1947: 50, Taf ) provides a good anchor for the date of their arrival (Вадаи, Кульчар 1984: 240, рис. 4. 7).

151 Sarmatians or Germans: Influence or Ethnic Presence? Old Problem New Hypotheses 151 Written sources do not specify on German presence at this territory in the period in question. The same is the situation with the find from Neyzats vault 306 which, according to Igor Khrapunov can belong to the time preceding Gothic campaigns to the North- Pontic coast (Храпунов 2011: 8). Presence of Germanic umbos was also recorded in a similarly authentic Sarmatian milieu in the barrows Sadovyy and Vysochino at the Lower Don region. Interestingly enough, in the latter case the shield together with silver vessels was hidden into a secret niche made in the wall of the grave which shows how valuable this object was for its owner (Щукин 2005: 72). The dating of these finds (stage B2, AD ) more or less coincides with the time of the horizon in question. In the Crimea and at the northern part of the Hungarian Plain, finds of Germanic character start to appear in clearly Sarmatian milieu at the same period. In Hungary authors of previous publications explained the presence of Vandalic objects with Sarmatian-Przeworsk neighbourship. However, a similar situation in a Sarmatian assemblage in the Crimea made us suggest that there could be another possible explanation. In both cases we deal with fresh immigrants. In neither of the cases are the Przeworsk features dominant. On the basis of the considerations represented above, a possible solution can be that at least part of both Hungarian and Crimean new Sarmatian population must had arrived from a Sarmatian-Germanic contact zone. Objects of Germanic (Przeworsk) character would more likely attest to an impact and not an actual presence of Germans in the Upper Tisza region. However, we should keep in mind that in the Crimea their sporadic presence is evidenced by single cremation burials in the cemeteries of the same period (Храпунов 2004: 141). In the third century a general crisis can be observed in the whole of the known world (Weltkrise) (Alföldi 1967). This is the period of the Hun, Gothic and Sassanid expansions, all of which had some kind of impact onto the peoples examined by us. Taking into consideration the phenomenon of the world crisis, it can be suggested that the simultaneous appearance of Germanic material in the Carpathian Basin and the Crimean peninsula is the consequence of the same historic process. For the deeper interpretation of this phenomenon we need further investigations. In this paper we would only like to draw attention to some aspects of it. Bibliography Alföldi A. A pesti oldal új urai; Aquincum a későrómai világban. // Budapest története I. Budapest az ókorban II. Budapest, Alföldi A. Studien zur Geschichte der Weltkriese des 3. Jahrhunderts nach Christus. Darmstadt, Bóna I. Szabolcs-Szatmár megye régészeti emlékei I // Szabolcs-Szatmár megye műemlékei I. Budapest, Cassius Dio. Roman History / tr. E. Cary. London; Cambridge; Massachusets, 1968 // uchicago.edu/thayer/e/roman/texts/cassius_dio (accessed: ). Gindele R., Istvánovits E. Die römerzeitliche Siedlung von Csengersima-Petea. Satu Mare, Godłowski K. Die Barbaren nördlich der Westkarpaten und das Karpatenbecken Einwanderungen, politische und militärische Kontakte // Specimina Nova dissertationum ex Instituto Historico Universitatis Quinqueecclesiensis de Iano Pannonio Nominatae Pécs, Historia Augusta. The Life of Marcus Aurelius // Historia_Augusta/Marcus_Aurelius/1*.html (accessed: ). Istvánovits E. Settlements of the Imperial Age in the Upper Tisza Region // Študijne Zvesti. 36. Nitra, Istvánovits E., Kulcsár V. Pajzsos temetkezések a Dunától keletre eső kárpát-medencei Barbaricumban // A nyíregyházi Jósa András Múzeum Évkönyve. XXX XXXII ( ). Nyíregyháza, Istvánovits E., Kulcsár V. New archaeological phenomena in the ethnical picture of Eastern Hungary after the Marcomannic-Sarmatian war // Specimina Nova dissertationum ex Instituto Historico Universitatis Quinqueecclesiensis de Iano Pannonio Nominatae. IX (1993). Pécs, 1994.

152 152 Eszter ISTVÁNOVITS, Valéria KULCSÁR Kobály J. A Przeworsk kultúrához tartozó harcossírok és fegyverleletek a Kárpátalján // Anyíregyházi Jósa András Múzeum Évkönyve. XXXIX XL ( ). Nyíregyháza, Nagy M. The Hasdingian Vandals in the Carpathian Basin. A preliminary report // Specimina Nova dissertationum ex Instituto Historico Universitatis Quinqueecclesiensis de Iano Pannonio Nominatae. Pécs, Németi J., Gindele R. Beiträge zur Geschichte des Careigegend im Jahrhundert n. Chr. // Acta Musei Porolissensis. XXI. Zalău, Olędzki M. The Upper Tisza Basin in the Roman Period. Remarks on settlement and cultural changes // Das mitteleuropäische Barbaricum und die Krise des römischen Reiches im 3. Jahrhundert / Spisy AÚ AV ČR. 12. Brno, Párducz M. Szarmatakori problémák // Antiquitas Hungarica I. Párducz M. Denkmäler der Sarmatenzeit Ungarns. III. Budapest, Prohászka P. Imperium et barbaricum. Kapcsolatrendszerek a Kárpát-medencében a római birodalom és a barbár népek között a római császár- és koranépvándorlás-korban (PhD thesis). Budapest, 2008 // (accessed: ). Schmidt L. Geschichte der Wandalen. München, Zoltai L. Die Hügelgräber der römischen Kaiserzeit in Hortobágy // Dissertationes Pannonicae. II / 11. Budapest, Вадаи А. Х., Кульчар В. К вопросу о так называемых сарматских пряжках // AAH. 36. Budapest, Русанова И. П. Культура карпатских курганов // Славяне и их соседи в конце I тысячелетия до н. э. первой половине I тысячелетия н. э. / Археология СССР. М., Храпунов И. Н. Этническая история Крыма в раннем железном веке / БИ VI. Храпунов И. Н. Склеп с погребениями III IV вв. н. э. из могильника Нейзац. Симферополь, Щукин М. Б. Готский путь. СПб., Эстер ИШТВАНОВИЧ, Валерия КУЛЬЧАР Сарматы или германцы: влияние или этническое присутствие? Старая проблема новая гипотеза Резюме Одна из давних дискуссионных проблем в изучении Барбарикума Карпатского бассейна это появление и этнический состав культурной группы в Верхнем Потисье. По свидетельству письменных источников, ко торые подтверждаются и археологически, вандальские группы осели в этом регионе во время или после Маркоманнских войн по соседству с сарматами (рис. 1). Памятники ранних вандалов связывают с воинскими погребениями-сожже ниями пшеворского облика. К сожа лению, дальнейшую историю вандалов на этой тер ритории проследить невозможно, так как го ризонт кремаций быстро исчезает, и позд нейших археологических признаков пшевор ского населения в северной части Алфёльда не остаётся. Венгерские и зарубежные исследо ватели традиционно пытаются иден ти фицировать их археологический материал III в. с так называемой группой «Гестеред Херпай Хортобадь». Памятники этой группы включают в себя сарматские курганные погребения, датирующиеся с наибольшей вероятностью периодом после середины III в., в каждом из которых обнаружены находки германского (пшеворско-

153 Sarmatians or Germans: Influence or Ethnic Presence? Old Problem New Hypotheses 153 го) характера. В числе последних обычно подчеркивается роль оружия. Согласно давнему общепринятому мнению, для сарматских комплексов присутствие щитов не характерно. Это относится и к наиболее ранним (период после Маркоманнских войн) сарматским погребениям Верхнего Потисья, население которого, по-видимому, соседствовало с пшеворцами, жив шими по ту сторону «ирано-германской гра ницы» (рис. 1). Ряд исследователей предполагает, что создателями упомянутой группы памятников являлись германцы или сарматы (аланы), смешавшиеся с германцами (Nagy 1993: , ; Prohászka 2008: 50 55; Bóna 1986: 63 66). Однако мы не располагаем непосредственными доказательствами в пользу присутствия германцев в северной части сарматского Барбарикума во второй половине конце III в. В то же время несомненно, что данная группа населения хотя бы частично представляет собой новую волну, новый археологический горизонт иранского населения (будь то роксоланы, аланы или другое сарматское племя или группа). Интересно отметить, что германские (пшеворские) элементы появляются в Крыму в этот же период. Нельзя считать совпадением, что одна из характерных германских находок из склепа 306 могильника Нейзац умбон щита типа Ikljær 3b, Jahn 7a (см.: Храпунов 2011: 8) тождественен находке из кургана II/10 могильника Хортобадь-Порошхат. Такие же умбоны известны в том же регионе сарматского расселения из погребения с типичным иранским обрядом (южная ориентировка, ингумация) в Ниредьхазе на ул. Арпад, а также из случайных находок Гестеред и Тисалёк- Разомпуста (Istvánovits, Kulcsár 1992: 55, 32, 50, 63). Наиболее известная находка этого типа умбон из Херпая. Схожая ситуация представляется нам в отношении находки из склепа 306 в Нейзаце, который, по мнению И. Н. Храпунова, принадлежит ко времени, предшествующему готским походам в Северное Причерноморье (Храпунов 2011: 8). В Крыму и в северной части Алфёльда находки германского характера появляются в чисто сарматском контексте более или менее одновременно. В Венгрии вандальские ве щи объясняли географической близостью сар матов и пшеворской группы в Верхнем По тисье. Однако схожая ситуация в сар матском комплексе в Крыму навела нас на мысль о том, что возможно и другое объяснение. В обоих случаях речь идет о недавно переселившихся мигрантах. Пшеворские элемен ты представлены в их материале слабо. На ос новании соображений, представленных выше, можно предположить, что обе сар мат ские группы крымская и «венгерская» хотя бы отчасти родом из какой-то сармато-германской контактной зоны. Вещи гер манского (пшеворского) характера говорят, ско рее всего, о влиянии, а не о собственно при сут ствии германцев в Верхнем Потисье. Можно предположить, что одновременное появление германского материала в отдельных регионах Кар патского бассейна и Крыма является послед ствием одного и того же исторического про цесса, на некоторые аспекты которого мы хо тели бы привлечь внимание в данной статье.

154 154 Michel KAZANSKI Water Routes from the Baltic to the Black Sea and Northern Barbarians in the Migration Period There were three well-known water routes from the Baltic Sea to the Pontos in the Roman period: by the Vistula and the Bug, by the Neman and the Berezina, and by the Dvina and the Dnieper (Kazanski 1992; Ščukin 2000). Here I will discuss the archaeological evidence concerning these routes in the Great Migration period (phase D in the chronology of the European Barbaricum, i. e. AD 360/ /510). 1 We will also study whether there are traces of northern barbarians from Scandinavia in these routes. The Vistula-Bug route The most familiar water way went from the southern Baltic Sea area along the Vistula and the Western Bug, and further to the basin of the Dniester and the Southern Bug; it is the famous Gothic road to the land of Oium (Jord. Get. 28), i. e. to the Ukrainian forest-steppe and steppe areas (on its location see: Щукин 2005: 149, pис. 52; Shchukin, Kazanski, Sharov 2006: 20, 35). It is well marked by many categories of archaeological finds from the Late Roman period (see for example: Kokowski 1995: ryc ; Ščukin 2000: fig. 3; Shchukin, Kazanski, Sharov 2006: 32, fig. 19; Mączyńska 2007: Karten 5 6; Mączyńska et al. 2007: Abb. 38, 40, 44; Andrzejowski, Prochowicz, Rakowski 2008: fig. 9). This route possibly started working normally when the Goths migrated by it in the late second century AD. Earlier, this area along the western Bug corresponded to the land of mutual fear of barbarian tribes in Germania and Sarmatia. Before the migration of the Goths, main contacts between the Baltic and the Pontic area laid either along the northern slope of the Carpathians or via the Neman and the Berezina or the Dvina and the Dnieper (Shchukin, Kazanski, Sharov 2006: 19). There is no written source to describe this route in the Great Migration period. However, the Vistula, as a part of the Amber Road, was certainly used in the Great Migration period as a connection between the Baltic Sea and the middle Danube, as rather numerous archaeological and numismatic finds suggested (for details see: Казанский 2010b: 21 24). Particularly, there is a famous treasure of Basonia, with ca. 300 kg of Baltic amber and ca. 30 kg of amber beads, discovered in this track (Schätze 1995: no ). The trade in amber probably continued in the fifth century. The concentration of the fifth century gold coins near the Vistula estuary suggests an important role played by this river in the system of international communications (see maps at: Maczyńska 2007: Karten 1 2; Mączyńska et al. 2007: Abb ). We can suppose that a branch of this route came by the upper Vistula and the Western Bug. In the Great migration period, the Western Bug connection is marked by engraved tongue-shaped beltends (fig. 1 2), distributed from the north to south, as far as the sources of the Western Bug (Kokowski 1999: fig. 3 4; Maczyńska 2007: Karte 7; Madyda- Legutko, 2011: mapa 23 24). Mapping of coins allows us to clarify the direction of the Vistula route, with one of its sections along the Vistula as far as its upper reaches and then to the south-east, to the upper reaches of the Western Bug and Volhynia (fig. 3) (Maczynska 1999: fig. 4; Kokowski 1999: fig. 9). In my point of view, artefacts from the Great Migration period in the Ukraine and the northern Black Sea area suggest that the water road along the Vistula and the Western Bug functioned and probably met in some way the water communication system in the Dniester, the Southern Bug, and the Dnieper basin. 2 1 I am grateful to Maxim Levada who helped me a lot with this paper. 2 Here we will not discuss the princely finds like Kachin or Bar with parallels in the Circum-Baltic area and in the middle Danube (Levada 2011). They reflect the princely fashion that, in my point of view, spread from Central Europe in northern and eastern directions. Something similar happened to some elements of the Hunnic princely culture that appeared as far as Scandinavia (Sösdala and Sjörup, see: Fabech ; Fabech 1991).

155 Water Routes from the Baltic to the Black Sea and Northern Barbarians in the Migration Period 155 Fig. 1. Distribution of the belt-ends Madyda 11 along the Vistula route (Madyda-Legutko 2011)

156 156 Michel KAZANSKI Fig. 2. Distribution of the belt-ends Madyda 12/2 along the Vistula route (Madyda-Legutko 2011)

157 Water Routes from the Baltic to the Black Sea and Northern Barbarians in the Migration Period 157 Fig. 3. Distribution of coins in the upper and middle Vistula basin in the Great Migration period (AD ). 1 AD ; 2 AD (1 Kokowski 1999; 2 Mączyńska 1999) Let us mention, first, the finds in the upper reaches of the Dniester and the Prut rivers. There is a strap-end (fig. 4. 2) discovered in the settlement of Ripnev II of the Dniester group of the Chernyakhov culture (Баран 1981: табл ). 3 Such endpieces are typical of the cultures of the Western Balts in Prussia and Germanics of the Wielbark culture from phase D1 according to the chronology of the European Barbaricum (AD 360/ /410) (Bitner-Wroblewska 1989: fig. 6, Carte 3; Nowakowski 1996: Taf ; Kokowski 1999: fig. 4; Prochowicz : ryc. 2; Madyda- Legutko 2011: mapa 22). Most likely, crossbow fibula of the type Prague- Wilkancy (fig. 4. 1) from the settlement Kodyn I of the Prague culture (Русанова, Тимощук 1984: рис ) was delivered to the upper Prut are by the Vistula route (Schulze-Dörrlamm 1985: ; Гавритухин 1989; Гавритухин 2005: 406). Such fibulae with distinctive flat bow 4 date back to the fifth century and were distributed in the south-eastern Baltic Sea area (see details and a catalogue of finds at: Медведев 2009). I can suppose that the fibula from Kodyn I is an evidence for some connections between the south-eastern Baltic Sea area and the Dniester area, which logi- 3 Most likely, the pendant with a scroll on the end from the settlement of Ripnev II (Баран 1981: рис ) originates from Central Europe (see parallels in Mĕcholupy and Gültlingen: Баран 1981: 119). It possibly belongs to the Prague layer of this site similarly to cast Byzantine brooch from the age of Justinian (Баран 1981: рис ; on the cast brooches see: Uenze 1992: ). 4 They should be separated from related brooches of the type Prague Pil viny, with twisted bow, widely distributed in Central Europe (see: Schulze-Dörrlamm 1985: ; Гавритухин 1989). Here I have not used the finds of Prague Pil viny brooches (e. g. in the settlements of Kolodeznyy Bugor and Kodyn: Третьяков 1974: рис ; Русанова, Тимощук 1984: pис ), because they could be brought from Central Europe rather than the southeastern Baltic Sea area.

158 158 Michel KAZANSKI Fig. 4. The Great Migration artefacts from the Vistula basin in the Dniester and the northern Black Sea area, and the northern elements in the Chernyakhov culture. 1 Kodyn I, dwelling 21; 2 Ripnev; 3 Al malyk-dere, burial 2/2001; 4 Tanais, barrow 1, burial 1 of 1960; 5 Skalistoye, burial 421; 6 7 Bashmachka; 8 10 Malaeşti, burial 3 (1 Гавритухин 2005; 2 Баран 1981; 3 Mączyńska, Urbaniak, Jakubczyk 2011; 4 Амброз 1969; 5 Веймарн, Айбабин 1983; 6 7 Shchukin, Kazanski, Sharov 2006; 9 10 Фёдоров 1960)

159 Water Routes from the Baltic to the Black Sea and Northern Barbarians in the Migration Period cally should be along water roads of the Vistula basin. Moreover, northern artefacts from the Great Migration period are also known in the east, as far as the Crimea and the lower Don area. Our attention should be paid to fibulae with returned foot, flat and sometimes ornamented bow, and knob on the head, similar to Oscar Almgren s type 163 they are probably derivatives of the latter (fig ). The main distribution zone of these fibulae coincides with the area of the Wielbark culture (Амброз 1966: табл ), where they have a relatively wide chronological frames (see for example Brest-Trishin: Кухаренко 1980: 46 47, табл. 8. 8a). Fibulae with returned foot and knob are known in the sites of the Chernyakhov culture (Kokowski 1995: 15, , ryc. 67, mapa 37) and in the northern Black Sea area (Kazanski 2002: 394, 400, 407). Anatoliy Ambroz has supposed that the distribution of these clasps is an evidence of the appearance of some post-gothic migrants, possibly the Gepids, in the northern Black Sea area (Амброз 1966: 67; Амброз 1969: 261). In the northern Black sea area, there are finds of such fibulae in grave no. 421 in the cemetery of Skalistoye (Веймарн, Айбабин 1993: 182, рис ), in the cemetery of Almalyk-Dere (Mączyńska, Urbaniak, Jakubczyk 2011: fig. 7. 6), and possibly in Tanais (Амброз 1969: 261, табл. 8. 1; Kazanski 2002: fig. 1.1) 5 (fig ). The brooch with ornamented flattened bow, similar to the Skalistoye find, discovered in vicinity of Thessalonica, could belong to the Great Migration period as well (Peschek : Abb. 1. 2). In the Great Migration period, the Chernyakhov area experienced the strengthening of some elements originated from the north, particularly in the Przeworsk culture (Kazanski 2009: ). 159 Among them, there is a cremation in urn, covered with shield boss, from period D1, in the cemetery of Malaeşti (Фёдоров 1960: 258, рис. 4. 4; 13. 1; 20. 2) (fig ). Typical Przeworsk ceramics from the top layer of Bashmachka settlement is worth paying attention too (fig ). However, this layer does not have an absolute date (for the site see: Смиленко, 1992). It is interesting to note that the early Great Migration period was the time of depopulation of Central and Northern Poland, i. e. the Przeworsk area (Godlowski, 1985: ), though some people remained there (Mączyńska 1999). Kaziemerz Godłowski has supposed that the demographic decline north of the Carpathians happened when a great deal of the population migrated to the south, to the area under the patronage of the Huns (Godlowski, 1985: ). 6 We can infer that some group of the Przeworsk people migrated to the Chernyakhov area (Shchukin, Kazanski, Sharov 2006: 145; Kazanski 2009: ). This is one of the possible reasons for the revival of the Vistula-Bug route in the Great Migration period. The Neman-Berezina route Archaeological data suggest the existence of the water route via the Neman Neris / Viliya / Wilia Berezina Dnieper as early as the Roman period (Амброз 1966: табл ; 18. 4; 23. 1; Kazanski 1992: 82 84, fig. 9). 7 The existence of this water route since extreme antiquity is evidences by hydronyms in the Neman and Dnieper basins. Such names as Svisloch Isloch and Berezina are mentioned in basins of both rivers. They duplicate each other: Isloch flows into the Berezina, a tributary of the Neman, though Svisloch with its sources near Isloch, flows into the Berezina, a tributary of the Dnieper (Погодин 1901: 12). This route is likely to function in the Great Migration period as well (fig. 5). 8 Especial activ- 5 Although, according to Ambroz, flattened bow of this brooch is a late feature, the context of the find is inexpressive. 6 Priscus of Panium informs that Attila s rule stretched on the islands in the Ocean (Hist. 8). If it is not a literary topos, thus we obtain an argument to infer the spread of the Hunnic power as far as the Baltic Sea. 7 The ancient estuary of the Neman could be the modern river Pregolya (Žulkus 2006: 17, fig. 1 2), though we do not know when the Neman changed its delta. Vladas Žulkus uses the background of the late mediaeval maps to compare the Neman with the Chronus in Ptolemaeus Geography (Žulkus 2006: 19). However, the late mediaeval maps are not the best argument to identify the rivers mentioned in the Geography. They often are nothing but reconstructions, so mediaeval Riga appears on the Chronus, which in this version is the same as the Western Dvina (e. g. Jacob Zigler s map of 1532, see: Савельева 1983: an insert). Generally, Ptolemaeus could describe not only large waterways but also smaller rivers known to him by some reason (Казанский 2010а: 124). 8 It could be reflected by Marcian, who informs that the river Rhudon in the Baltic area and the Borysthenes begin in the Alanic mountains, i. e. their sources are not far from each other (Periplus Maris Exteri 2. 39). We can possibly identify the Rhudon with the Neman according to Ptolemaeus ethnic and geographical data (see: Казанский 2010а: 126), but this is not the only version.

160 160 Michel KAZANSKI Fig. 5. The Neman-Berezina route in the Great Migration period. 1 Veliuona; 2 Kaunas; 3 Kašetos; 4 Grodno-Grodnichanka; 5 Gury; 6 Labenshchina; 7 Kamenevo-2; 8 Babina Gora; 9 Grigor yevka Chernechiy Les; 10 Lashchova; 11 Shul govka. Conventional signs: 1 brooches of the West Baltic tradition (outside the Baltic Sea area); 2 belt-end of the Scandinavian tradition; 3 coin finds ity on the Neman in the Great Migration period is indicated by the coin finds (fig. 5. 3): a treasure with a coin of Honorius (AD ) in Kaunas (Michelbertas 1972: no. 20), a coin of Valentinia nus I or II in Kašetos (Michelbertas 1972: no. 21), and a hoard with coins of Valens (AD ) in Veliuona (Michelbertas 1972: no. 66); finally, we should mention a coin of Theodosius (AD ) discovered in Grodno-Grodnichanka, in the basin of the Naman as well (Поболь 2001: 77). In the Great Migration period, a series of Baltic artefacts appeared in the Dnieper area (Shchukin, Kazanski, Sharov 2006: 145; Kazanski 2009: ; Казанский 2010b: ); their mapping suggests that they most likely penetrated there by the Neman-Berezina route (fig. 5. 1). There are crossbow fibulae with spade-shaped foot of Anna Bitner-Wroblewska s type I, from the first and second stages of the Western Baltic antiquities, i. e. AD (for the West Baltic chronology see:

161 Water Routes from the Baltic to the Black Sea and Northern Barbarians in the Migration Period Bitner-Wroblewska 2001: 14 19, ), better known among the antiquities if the Western Balts in the former East Prussia (Bitner-Wroblewska 2001: 59 65, fig. 11). According to the find (fig. 7. 6) from Belorussian Gury (Митрофанов 1978: рис ), these clasps spread by the Neman Berezina Dnieper water route. In the Dnieper area, they are discovered in Slavonic settlements of Kamenevo-2 (e. g.: Терпиловский 2004: рис ), Gri gor yevka Chernechiy Lis (Петрашенко, Циндровська 1999: рис. 1. 8), near Kanev and Pereyaslav (Kazanski 2000b: 427, fig ), and in the Chernyakhov settlement of Lashchova in vicinity of modern Cherkassy 9 (fig ). Another northern fibula of type Schönwarling is discovered in Babina Gora (fig ), on the right bank of the Dnieper (Kazanski 2000b: 428, fig. 6. 5). These clasps appeared during phase 1 (AD 350/ /400), and especially in phase 2 (AD 375/ ), according to the Western Baltic timeline of the Great Migration period, in Prussia and Mazovia (Bitner-Wroblewska 2001: 34 40, fig. 3). A fibula of type Sensburg-Mrągowo (Bitner- Wróblewska 2001: 79 83) originates from Shul govka (fig ) near the Dnieper rapids (Казанский 1999: ). These clasps are known in Öland and Bornholm, near the Neman estuary, in the basin of the Gauja, and in Masuria (Bitner-Wróblewska 2001: fig. 15). Their precise chronology is hardly determinable. In Bornholm, such brooches appeared in the burial context of the Early Germanic Iron Age, which nearly corresponded to the period from the mid-fifth to the early sixth century (Bitner-Wroblewska 2001: 87). There is a fibula of type Dollkeim-Kovrovo from some place in the middle Dnieper area (fig ) that was stored in the National Museum of the History of the Ukraine (Kiev) before the 161 Second World War and then found itself in the Archaeological Museum in Kraków (Kazanski 2000b: fig. 6. 7). 10 These clasps are derivatives from the type Schönwarling and are especially studied by Anna Bitner-Wróblewska. In her typology, this brooch is a transitional type with the catch plate shorter than the foot-plate. These fibulae are often discovered in former East Prussia, especially in Sambia, also in Gotland and Öland, and one find is in the lower Neman area (Bitner-Wróblewska 2001: 41 52, fig. 6). In Bitner-Wróblewska s opinion, the transitional variant of fibulae appeared in the land of the Balts as early as the Great Migration period, though in Scandinavia they belonged to period VI. 2 of Swedish timeline, i. e. the second half of the fifth and early sixth century (Bitner-Wróblewska 2001: 50 52). According to the Gotland materials, the transitional variant of fibulae was most typical of phase D2b of local chronology, i. e. AD (Rundkvist 2003: fig. 2. 6). Classical Dollkeim- Kovrovo fibulae with equal length of foot-plate and catch-plate, like the find from Zaostrov ye-1, belong, according to the Polish researcher, to stages 2 4 of the West Baltic timeline, i. e. AD 375/ /530 (Bitner-Wróblewska 2001: 50 52, рl. 59). A pair of classical Dollkeim-Kovrovo fibulae is discovered in treasure of Frauienberg / Frombork with a coin of Theodosius II from AD 430, so they confirm the date suggested above (Bitner- Wróblewska 2001: pl ). In the area of the Neman-Berezina route, there were more exotic artefacts. The top layer of the fortified settlement of Labenshchina (Tushemlya culture), located at a tributary of the Svisloch in the Berezina basin, contained strap-end of type Stora Hästnüs (fig. 7. 5) from Gotland (Митрофанов 1978: рис ). 11 Scandinavian origin of these 9 Fyodor Androshchuk discovered a bronze brooch, 9.7 cm long, in 1997 on the Chernyakhov settlement at the modern village of Lashchova in Tal noye district of Cherkassy oblast (administrative area). I am grateful to Androshchuk and Levada for this information and drawing. It should be a local variant of fibulae with spade-shaped foot, because its spring-holder differs from the Baltic clasps. Another fibula of the type originates from illegal excavations in the Ukraine, so its photographs are placed on a web site of antiquity dealers. 10 I am very grateful to Kazimierz Godłowski for the possibility to get acquainted with this fibula during my trip to Kraków in 1994 upon his invitation. 11 The Labenshchina strap-end is discovered close to an openwork appliqué badge (Митрофанов 1978: рис ). As Anatoliy Aleksandrov (Pskov) kindly informs, this badge meets with parallels in the thirteenth and fourteenth century antiquities in Novgorod, being among the fittings of oriental (Tatar?) origin. Aleksandrov supposes that the strap-end in question could be a fitting of the same bag (his personal message of 21 March 2013). Although there is no mediaeval layer in Labenshchina, this oriental bag could be brought there by chance. I think that Aleksandrov s interpretation is possible, though it is necessary to admit that the Novgorod find cited by him does not contain strap-ends of the type of interest. Therefore, the finds in Labenshchina, namely the strap-end and the mediaeval bag appliqué, belong to different periods, perhaps, despite they were discovered together.

162 162 Michel KAZANSKI Fig. 6. West Baltic and Scandinavian artefacts distributed in the Neman-Berezina route and their parallels (1 4). 1 Kuninguste; 2 3 Stora Hästnäs; 4 Gotland; 5 Labenshchina; 6 Gury; 7 Grigor yevka-chernechiy Les; 8 9 Kamenevo; 10 Lashchova (drawing: Maxim Levada); 11, 14 Dnieper area (14 photo: the au thor); 12 Babina Gora; 13 Shul govka (1 4 Quast 2004; 5 6 Митрофанов 1978; 7 Петрашенко, Циндровська 1999; 8 9 Терпиловский 2004; Kazanski 2000)

163 Water Routes from the Baltic to the Black Sea and Northern Barbarians in the Migration Period strap-ends (for the examples in the Baltic area see: fig ) suggested by Dieter Quast seems beyond any doubts (Quast 2004: 263). 12 It is demonstrative that in the Great Migration period there were many Scandinavian elements at the beginning of the route under analysis: from the south coast of the Curonian coast to the ancient bay north of the modern city Elbląg. Let me remind that it was the time when the Neman estuary could be the Pregolya (see above). Primarily, there are prestigious artefacts of princely and warrior culture of Scandinavian and Western European origin, including neck-rings Andersson R 300 with overlapping ends (Andersson 1995: 946) from Hammersdorf / Mloteczno (fig. 7. 6) (for them see: Кулаков 2003: ; Эпоха 2007: IV. 1; IV. 4. б; Cieśliński 2008: ; Казанский 2010b: 53), 13 parade bladed weapons decorated in the first Germanic animal style from warrior elite burials in the cemetery of Warnikam / Pervomayskoye (fig. 7. 8, 10) (Кулаков 1997; Kulakov 1997; Hilberg 2009: ), rather numerous scramasaxes with narrow pointed tip and blood-grooves on the blade (Kazanski, Mastykova 2005: 123), details of horse harness decorated in animal style, with anthropomorphic elements, like saddle- and bridle-plates in grave 335 of Mitino cemetery (fig. 7. 7, 9) (Скворцов 2010: ), or parts of horse-harness in the first Germanic animal style in a grave in the cemetery of Warengen / Kotel nikovo (fig. 7, 5) (Hilberg 2009: 331, Abb ). In this area, there also are finds of women s attire of Scandinavian origin such as southern Scandinavian relief brooches with rectangular headplate in grave 13 of the cemetery of Detlevsruh / Pesochnoye (fig. 7. 1) in Sambia (Hilberg 2009: , Abb. 6. 1, Taf ; cf.: Sjøvold 1993: 20 30, pl ) and in grave 9А in the cemetery of Schakaulack / Malaya Lipovka (fig. 7. 2) (Hilberg 2009: Abb. 6. 4c, Taf ; cf. Sjøvold : pl. 32, no. 11), relief three-rayed brooch in the cemetery of Hünenberg / Dobroye (fig. 7. 3) in the Sambian peninsula (stray find: Кулаков 1990: 205, рис. 2. 5; Hilberg 2009: , Taf ; cf.: Sjovold 1993: 52 53, fig. 17, рl. 30. S27; 31. D15), and a brooch similar to Scandinavian cross-like clasps in the cemetery of Aukštkiemiai / Oberhof (fig. 7. 4) (Hilberg 2009: Taf ; cf.: Reichstein 1975: Taf ; ). In the fifth century, gold Romano-Byzantine coins concentrated in the Vistula estuary and bay, where the route under analysis started (fig. 8), though they were few in the Vistula basin. The route via the Neman and the Berezina, bypassing the basins of the Vistula, the Oder and the Danube, was probably related to certain political events in Central Europe. From time to time, there were violent conflicts that temporarily interrupted the network of Trans-European connections, for example during the invasion of the Huns and immediately after the death of Attila. Northern barbarians possibly participated in this skirting manoeuvre. The Dvina-Dnieper route One more thing to discuss is the possible route taken by the Scandinavians to go to the forest zone of Eastern Europe along the Gulf of Riga and the Western Dvina, known from the Late Mediaeval Scandinavian narratives, such as Gutasaga (Сыромятников 1892). 14 This route existed already in the Roman period (Kazanski 1992: 82 84, fig. 9) and at least until ca. AD 400, as it appeared from the archaeological materials and particularly the finds of the circle related to the treasure in Sukhonosivka (Левада 2010: , рис , 22, 25). In the period of interest, the lower Western Dvina basin was occupied by the East Baltic cultures: there were flat cemeteries of inhumation graves closer to the Western Dvina estuary, though 12 There is a Scandinavian bracteate of the Dreigötter-Brakteaten group, formerly considered the find from the Berezina, accompanied with coins from AD , , , (Axboe 1981: 86; Axboe 1994: 72 73, Appendix C), but today attributed as a part of the Polish treasure in Zagórzyn (Pesch 2011: Tab ). 13 These neck-rings of Scandinavian type appeared in the Roman period in Scandinavia and northern Germany and thence spread to the Baltic Sea area (Moora 1938: ; Lincke 1939; Аун 1992: ; Abegg-Wigg 2008: 31 34). In the eastern Baltic Sea area, there often were their local derivatives, which still reflected a Scandinavian- Southern-Baltic fashion. These neck-rings existed throughout the Great Migration period and even in the seventh, eighth, and ninth centuries (e. g. in grave 319 in Anduliai: Lincke 1939: Abb. 9). 14 Gutasaga forms the background for the conclusion that the Scandinavians migrated from Gotland to the Baltic Sea area in AD (Nerman 1929: 11). Although there are numerous finds of Scandinavian artefacts from the Great Migration period in the eastern Baltic Sea area indicating the presence of northern barbarians (Quast 2004; Казанский 2010: ; Казанский 2010с), indisputable sites of permanent Scandinavian settlements from the fifth and sixth centuries in the eastern shore of the Baltic Sea have not been discovered so far.

164 164 Michel KAZANSKI Fig. 7. Artefacts of the Scandinavian tradition in Sambia, Natangia, and Curonian coast. 1 Detlevsruh / Pesochnoye; 2 Schakaulack / Malaya Lipovka; 3 Rantau-Hünenberg / Dobroye; 4 Aukštkiemiai / Oberhof; 5 Warengen / Kotel nikovo; 6 Hammersdorf / Mloteczno; 7, 9 Mitino, grave 335; 8 Warnikam / Pervomayskoye, grave 4; 10 Warnikam / Pervomayskoye, grave 1. (1 6, 8, 10 Hilberg 2009; 7, 9 Скворцов 2010) upstream there were both flat cemeteries and barrows surrounded by stone circles and inhumations; fortified and unfortified settlements are recorded everywhere. The upper and middle Western Dvina areas are occupied by the sites of the Tushemlya or Tushemlya-Bantserovshchina culture (for the short review of this situation see: Казанский 2010b: , with bibliography). A population that created the flat cemetery of Pląvniekkalns (fig. 9. 2) appeared in the fifth century on the left bank of the Western Dvina, near its estuary, in the area that was depopulated in the Roman period (Radiņš 2006: 83, fig. 2. 3). It could be related with the revival of the river and sea routes near the Western Dvina and the Gauja estuaries in the Great Migration period (Казанский 2010b: 103). The probable route somewhere along the southern shore of the Gulf of Riga, with a certain relation to the Western Dvina and Gauja estuaries, is indicated by finds of early Byzantine coins (fig. 9. 1, 3 4). Apart from a doubtful treasure of Jaunzemii in the Gauja basin, there are gold solidus of Anastasios (AD ), discovered in Engure, on the sea shore west of the Dvina estuary, and bronze coin of Justinian I, the only in the Baltic Sea area, discovered in Razas, on the right bank of the lower Western Dvina (Казанский 2010b: 103). The finds of the fibulae of the West Baltic origin in the lower course of the Western Dvina show that there still were some contacts via this route

165 Fig. 8. The finds of the fifth century solidi near the Vistula estuary. 1 the first half of the fifth century AD: A coins from AD /7; B coins with non-specified date; C coin hoards. 2 the second half of the fifth century AD: A coins from AD 455/7 491; B coins from AD ; C coins of non-specified date; D coin hoards (Mączyńska 2007)

166 166 Michel KAZANSKI Fig. 9. The Western Dvina route in the Great Migration period. I coin finds; II West Baltic brooches. 1 Engure; 2 Pląvniekkalns; 3 Razas; 4 Braslav in the Great Migration period. Demonstrative are also star-footed brooches of types III and VII (in Pląvniekkalns, see: Bitner-Wróbliewska 2001: , fig. 11). The map of these finds indicates that type III most likely penetrated from the lower Neman basin and the Curonian coast by the sea. Type VII is more characteristic of the south-western Finland, northern and south-eastern Estonia; it is also known in the cemetery of Maudžiorai in Žemaitija (Bitner-Wróbliewska 2001: , fig. 11). It is still possible that brooches of type VII were brought to the Western Dvina by inner routes in the Baltic area, similarly to the Dollkeim- Kovrovo type (in Pląvniekkalns, Jaunjelgava, and Kunci, see: Bitner-Wróbliewska 2001: 179, fig. 7), Letto-Lithuanian variant of fibulae Sensburg- Mrągowo (in Jēkabris and Krustpils Mantas, see: Bitner-Wróblewska 2001: 196, fig. 16), and drinking horns or their details from the fifth and sixth century. 15 It is considered that the fibulae with rings on the body from period Е (Bitner-Wróblewska 2006: 11, fig. 7) were distributed along the Eastern Baltic sea area from Prussia and the Curonian coast, most likely by the sea, as Harri Moora has supposed (Moora 1938: ; Аун 1992: 140; Кулаков 2003: рис. 91). Brooches with rings on the body are recorded in the cemetery of Pląvniekkalns at the Western Dvina estuary (Moora 1929: Taf ; Graudonis 1974: att ). They are actually numerous in the cemeteries in Sambia and Curonian coast, though these fibulae also exist in the northern Lithuania, in the area between the upper Venta and Mūša, as well as in Zemgale, for example in the cemetery of Ķŭri on the Tērvete. These finds 15 Although there are such artefacts in Uppland (the mid-sixth century find from Hade, see especially: Magnus 1999: ) and in the Western Dvina basin (in Kentes, Lejasbitēne, and Lielpuderi, see: Andrzejowski 1991: no. 129, 166, 170), the chronology of Latvian finds requires clarification.

167 Water Routes from the Baltic to the Black Sea and Northern Barbarians in the Migration Period 167 Fig. 10. Artefacts of the East Baltic and western origin in the area of the Tushemlya culture. 1 6 Klishina, a hoard; 7 8, 10, Demidovka; 9 Nikodimovo; 11 Bliznaki (1 6 Шадыра 2006; 7 8, Шмидт 2003; 9 Седин 2011)

168 168 Michel KAZANSKI Fig. 11. Artefacts of western and southern origin in the area of the Tushemlya culture. 1 Khotyshcha; 2 3 unknown findspot; 4 Blučina; 5 6 Borki; 7 Smolensk oblast ; 8 Chapayevka; 9 10 Glotasitz; 11 Landirano; Demidovka (1 6 Амброз 1970; 7 Эпоха 2007; 8 Приходнюк 1998; 9 10 Gläser 2008; 11 Bierbrauer 1975; Шмидт 1960)

169 Water Routes from the Baltic to the Black Sea and Northern Barbarians in the Migration Period suggest that the sea route was not the only path to deliver the fibulae with rings on the body into the western Dvina estuary. North of the Western Dvina, these brooches sometimes appear at the river Gauja (the cemetery of Strante), and later appeared in Estonia and Finland (for details and bibliography of these finds see: Казанский 2010b: 56). Latvian archaeologists consider that the Western Dvina route was destabilized in the middle of the first millennium AD (The Selonians 2007: 246). However, I think that the contacts by this river continued as far as a certain place. A hoard of Braslav (Поболь 2001: 77) with gold coin of Valens (AD ), discovered in the Dvina area, not far from modern borderline between Byelorussia and Latvia, marks the easternmost point of the river road via the Western Dvina. However, Western Baltic imports did not come upstream the Dvina, to the area of the Tushemlya culture. There probably were reasons why the population that created the Tushemlya sites in the Western Dvina area did not let the owners of West Baltic artefacts go through their territory. Similar was the situation in the Roman period: the goods marking the contacts between the Baltic Sea area and the Volga and Oka basin did not come to the upper Dnieper area (Kazanski 1992: fig. 11). The finds of Scandinavian origin are also absent in the Western Dvina so far. It is interesting to note that the area of the Tushemlya sites on the Dvina received artefacts typical of the Eastern (Letto-Lithuanian) Balts, such as spears Kazakevičius II (fig ), horse bits with three-partite mouth-pieces (fig ), wire bracelet-like temple pendants (fig , 7), diadems with spiral beads (fig ), coiled finger-rings (fig ), shepherd s staff pins (fig ), hooked spurs (fig ), etc. (Kazanski 2000b: ; Kazanski 2000a; Казанский 2010b: 104). Although the contacts with some East Baltic groups probably continued, the route via the Dvina did not function in the Tushemlya area. In the Great Migration period, a series of southern imports was recorded in the Tushemlya area, for example grey-clay wheel-made (Chernyakhov?) vessel and composite belt of the Kapulovka type 169 (fig ) in Demidovka (Шмидт 1970: рис ; 4. 2; for such belts see: Kazanski 1993: ). Form the mid-fifth century onwards, there were artefacts in the Danube tradition but discovered in the upper Dnieper area, particularly radiate-headed brooches (fig ) and belt fittings (fig , 2, 7), pyramidal and adze-shaped fire-lighters (fig ) or piercers (fig ), etc. (Казанский 2010b: ; see also: Эпоха 2007: III. 2, III. 5; on the adze-shaped fire-lighters or piercers see: Schuster 2011). The Danube traditions are especially pronounced in the details of attire. Spiral decoration on the goods from Demidovka and Khotyshchy (fig ) has parallels among the buckles from the middle and the second half of the fifth century in Blučina (fig ), Bistreţ, Rommersheim (Or 2000: no ; Marcu 1987: fig. 4. 1; Kessler, Schnellerkamp 1933: Abb. 5. 1), as well on the scabbard from East Germanic parade sword in the princely grave in Taurapilas (Tautavicius 1981: pav. 12), dated to the middle and the second half of the fifth century by the belt buckles of type Gava (on their date see: Tejral 1988: 279), and, finally, on some Central European fibulae from the same period (Svoboda 1965: t ; Merowingerzeit 1995: Taf. 8). Motif of two birds heads looking in opposite directions on the buckle from Demidovka (fig ) is known in the East Germanic ornaments from the second half of the fifth century onwards (for example on an Italian brooch from Desana: Bierbrauer 1975: Taf ; today it is dated back to the period of Odoacer and not to the period of the Ostrogothic kingdom: Menke 1986: 262). Decorative motif with two birds or beasts heads facing each other on appliqués and buckle panels (fig , 7) is known on the belt fittings in the upper Dnieper area and in cemeteries in the middle Oka area (for example: Амброз 1970: рис. 1. 4; Эпоха 2007: III. 5. 1). It also has Italian-Ostrogothic and Danube parallels (fig ) (for example: Bierbrauer 1975: Taf ; Gläser 2008: Abb. 4 5). 16 Lattice pattern on belt fittings from Smolensk and the Oka areas (fig ) (Амброз 1970: рис. 1. 3, 4, 6; Эпоха 2007: III. 5. 1) is present on various artefacts from the mid-fifth to the late sixth century (Vallet 1988: 52; 16 One buckle of the type is known in the middle Dnieper area as well, and namely in Chapayevka (fig ) (Приходнюк 1998: рис ), which makes another indication of the role played by the Dnieper water road in distribution of such artefacts. It is still possible that the sixth century Black Sea brooch of the late variant Gurzuf which originated from somewhere in modern Latvia (Kühn 1974: Taf ) was also brought to the Baltic area by the Dnieper-Dvina route.

170 170 Michel KAZANSKI Tejral 1988: Abb ; Tejral 2005: Abb. 3. D. 23; I Goti 1994: fig. III. 23). 17 These artefacts from the Pontic-Danubian area could be arguments for the suggestion that in the late fourth and first half of the fifth century the upper Dnieper area received groups of southern and western populations: the Veneti, their Hunnic allies, the Goths defeated by the Huns, and others. In this period, destructions probably caused by war are recorded in the fortified settlement of Demidovka (for details see: Фурасьев 1996; Казанский 1999а; Kazanski 2000b). It looks like these events destabilized the route via the Dnieper and the Dvina. *** Thus, there are few archaeological materials supplying evidence for the river roads between the Baltic Sea and the Black Sea that worked in the Roman period still existed in the Great Migration period. First of all, there were routes via the Vistula and the Western Bug, and via the Neman, the Wilia, the Berezina and the Dnieper. The development of the route via the Vistula and the Bug could be related with the migration of a part of Central European population to the south-east. The appearance of several Scandinavian artefacts in the Neman Bere zan route marks the appearance of persons originated from the Baltic area. According to the distribution of coins and Western Balt artefacts, the water route along the Western Dvina to the upper Dnieper was broken in the fifth century, so only its section in the lower Dnieper was used. Most likely, it was because of some population groups penetrated into the upper Dnieper area from the south or south-west. Bibliography Abegg-Wigg A. Germanic and Roman-provincial symbols of power Selected finds from the aristocratic burials of Neudorf-Bornstein // The Turbulent Epoch. New materials from the Late Roman Period and the Migration Period. I. Lublin, Andersson K. Romartida guldsmide i Norden. III. Uppsala, Andrzejowski J. Okucia rogów do picia z młodszego okreus prezdrzymskiego i okresu wpływów rzymskich w Europie Środkowej i północnej // Materilay starozytne i wczesnosredniowieczne Andrzejowski J., Prochowicz R., Rakowski T. Brooches Type Gródek 47 and their dating // The Turbulent Epoch. New materials from the Late Roman Period and the Migration Period. I. Lublin, Axboe M. The Scandinavian gold bracteates. Studies on their manufacture and regional variations // Acta Archaeologica Axboe M. Gudme and Gold Bracteates // The Archaeology of Gudme and Lundeborg. København, Bierbrauer V. Die ostgotischen Grab- und Schatzfunde in Italien. Spoleto, Bitner-Wroblewska A. Elementy bałtyjskie w kulturze wielbarskiej // Kultura wielbarska w młodszym okresie rzymskiem. 2. Lublin, Bitner-Wróblewska A. From Samland to Rogaland. East-West connections in the Baltic basin during the Early Migration Period. Warszawa, Bitner-Wróblewska A. The Key Problems of Late Migration Period in the Balt Lands // Transformatio Mundi. The transition from the Late Migration Period to the Early Viking Age in the East Baltic. Kaunas, The Danube wave of influences into the zone of the Tushemlya culture possibly brought the axe (fig ) similar to West Germanic battle-axes Hübener A (Hübener 1980: 72, Abb , 44), discovered in the fortified settlement of Demidovka (Шмидт 2003: табл ). Such axes armed the Merovingian warriors, as well as those in England, northern Italy, middle Danube, and northern Balkans (Hübener 1980: 91 96; Kazanski 2000a; 2001). However, it seems that the body of the Demidovka axe is more solid than of its Germanic analogies.

171 Water Routes from the Baltic to the Black Sea and Northern Barbarians in the Migration Period 171 Cieśliński A. Die spätrömische Kaiserzeit und die frühe Völkerwanderungszeit an der Passarge / Pasłę ka // The Turbulent Epoch. New materials from the Late Roman Period and the Migration Period. I. Lublin, Fabech C. Sjörup- an Old Problem in a New Light // Meddelanden från Lunds universitets historiska museum Fabech C. Booty Sacrifices in Southern Scandinavia: A Reassessment // Sacred and Profane. Oxford, Gläser F. Die Goten und der Arianismus im Alpen-Adria-Raum // Rom und die Barbaren. Roma, Godlowski K. Przemiany kulturowe i osadnicze w południowej i środkowej Polsce w młodszym okresie przedrzymskim i w okresie rzymskim. Wroclaw; Warszawa; Kraków; Gdansk; Lódz, I Goti. Milano, Graudonis J. Agrais dzelzs laikmets // Latvijas PSR Archeologija. III. Riga, Hilberg V. Masurische Bügelfibeln. Studien zu den Fernbeziehungen der völkerwanderungszeitlichen Brandgräberfelder von Daumern und Kellaren. Daumen und Kellaren Tumiany i Kielary. 2. Neumünster, Hübener W. Eine Studie zu den Beilwaffen der Merowingerzeit // Zeitschrift für Archäologie des Mittelalters Kazanski M. Les arctoi gentes et l empire d Hermanaric // Germania /1. Kazanski M. Les armes baltes et occidentales dans la zone forestière de l Europe orientale à l époque des Grandes Migrations // Archaeologia Baltica. 2000a. 4. Kazanski M. La zone forestière de la Russie et l Europe centrale à la fin de l époque des Grandes Migrations // Die spätrömische Kaiserzeit und die frühe Völkerwanderungszeit in Mittel- und Osteuropa. Lodz, 2000b. Kazanski M. Les antiquités germaniques de l époque romaine tardive en Crimée et dans la région de la mer d Azov // Ancient West & East I/2. Kazanski M. L époque hunnique au Nord de la mer Noire // Kazanski M. Archéologie des peuples barbares. Bucarest; Brăila, Kazanski M., Mastykova A. Les contacts entre la Gaule du Nord et la côte sud-est de la mer Baltique durant l époque des Grandes Migrations et au début de l époque mérovingienne // Voies d eau, commerce et artisanat en Gaule mérovingienne. Namur, Kessler A., Schnellerkamp W. Ein frühmerowingisches Grab bei Rommersheim (Eichloch) in Rheinhessen // Mainzer Zeitschrift Kokowski A. Grupa masłomecka. Z badań nad przemianami kultury gotów w młodszym okresie rzymskim. Lublin, Kokowski A. La fin de la civilisation de Wielbark et ses conséquences // L Occident romain et l Europe centrale au début de l époque des Grandes Migrations. Brno, Kühn H. Die germanischen Bügelfibeln der Völkerwanderungszeit in Süddeutschland. Graz, Kulakov V. Gräber pruzzischer Stammesführer aus Warnikam // Eurasia Antiqua Levada M. To Europe via the Crimea: on possible migration routes of the northern people in the Great Migration period // Inter Ambo Maria: Contacts between Scandinavia and the Crimea in the Roman Period. Kristiansand; Simferopol, Lincke B. Eine baltische Halsringform der Völkerwanderungszeit // Prähistorische Zeitschrift (1 2). Mączyńska M. La fin de la culture de Przeworsk // L Occident romain et l Europe centrale au début de l époque des Grandes Migrations. Brno, Mączyńska M. Pommern in der Völkerwanderungszeit 20 Jahre nach dem Aufsatz von Kazimierz Godłowski // Barbaren im Wandel. Beiträge zur Kultur- und Identitätsumbildung in der Völkerwanderungszeit. Brno, 2007.

172 172 Michel KAZANSKI Mączyńska M., Andrzejowski J., Gorecki J., Hammer P., Pawlikowski M., Rudnicka D., Voß H.-U. Der frühvölkerwanderungszeitliche Hortfund aus Łubliana, Kreis Kościerzyna (Pommern) / Bericht der Römisch-Germanischen Kommission. 90. Mainz, Mączyńska M., Urbaniak A., Jakubczyk I. The Early Medieval Cemetery of Almalyk-Dere near the Foot of Mangup // Inter Ambo Maria: Contacts between Scandinavia and the Crimea in the Roman Period. Kristiansand; Simferopol, Madyda-Legutko R. Studia nad zróznicowaniem metalowych części pasów w kulturze przeworskiea. Okucia końca pasa. Kraków, Magnus B. The assemblage from Hade in Gästrikland and its relevance for the chronology of the late Migration Period in eastern Sweden // The Peace of Change. Studies in Early Medieval Chronology. Oxford, Marcu D. Un mormînt de îhumatie descoperit la Bistret, jud. Dolj // Studii şi Cercetare de Istorie Veche /2. Menke M. Archäologische Befunde zu Ostgoten des 5. Jahrhunderts in der Zone nordwärts der Alpen // Peregrinatio Gothica / Archaeolоgia Baltica. VII. Lodz, Merowingerzeit. Die Altertümer im Museum für Vor- und Frühgeschichte, Berlin. Mainz, Michelbertas M. Prekubiniai ryšiai su Romo imperija // Lietuvos gyventoju prekybiniai ryšiai I XIII a. Vilnius, Moora H. Die Eisenzeit in Lettland bis etwa 500 N. Chr. I. Die Funde / Õpetatud Eesti Seltsi Toimetused. XXV. Tartu, Moora H. Die Eisenzeit in Lettland bis etwa 500 N. Chr. II. Analyse / Õpetatud Eesti Seltsi Toimetused. XXIX. Tartu, Nerman B. Die Verbindungen zwischen Skandinavien und dem Ostbaltikum in der jüngeren Eisenzeit. Stockholm, Nowakowski W. Das Samland in der römischen Kaiserzeit und seine Verbindungen mit dem römischen Reich und der barbarischen Welt. Marburg; Warszawa, L Or des princes barbares. Du Caucase à la Gaule, Ve siècle après J.-C. Paris, Pesch A. Gold Bracteats and Female Burials // Weibliche Eliten in der Frühgeschichte. Mainz, Peschek C. Zur Südausbreitung der Fibel mit ungesclagenem Fuss // Prähistorische Zeitchrift Prochowicz R. Znalezisko dzibowatego okucia końca pasa z Przeradowa, pow. Makowski // Wiadomości Archeologiczne Quast D. Ein scandinavisches Spathascheidenmundblech der Völkerwanderungszeit aus Pikkjärve (Põlvamaa, Estland) // Jahrbuch des RGZM Radiņš A. Lower Daugava Area in the 1 st 11 th Century. Ethnic, Economic, social and Political Change on the Question of Activity along Daugava Waterway // Transformatio mundi. The Transition from the Late Migration Period to the Early Viking Age in the East Baltic. Kaunas, Reichstein J. Die kreuzförmige Fibel. Zur Chornologie der späten römischen Kaiserzeit und der Völkerwanderungszeit in Scandinavien, auf dem Kontinent und in England / Offa-Bücher. 34. Neumünster, Rundkvist M. Barshalder 1. A cemetery in Grötlingbo and Fide pareishes, Gotland, Sweden, c. A. D Excavations and finds Stockholm, Schätze der Ostgoten. Stuttgart, Schulze-Dörrlamm M. Romanisch oder Germanisch? Untersuchungen zu den Armbrust- und Bügelknopffibeln des 5. Und 6. Jahrhunderts n. Chr. Aus dem Gebieten westlich der Rheins und südlisch der Donau // Jahrbuch des RGZM Schuster J. Przekłuwacze typu Dresden-Dobritz / Źrniki Wielkie. Uwagi na temat nadrzędzi z późnego okresu rzymskiegoi wczesnej fazy okresu wędrówek ludów // Wiadomości Archeologiczne

173 Water Routes from the Baltic to the Black Sea and Northern Barbarians in the Migration Period 173 Ščukin M. Three Ways of the Contacts between the Baltic and the Black Sea Littorals in the Roman Period // Die spätrömische Kaiserzeit und die frühe Völkerwanderungszeit in Mittel- und Osteuropa. Lodz, Shchukin M., Kazanski M., Sharov O. Des Goths aux Huns: Le Nord de la mer Noire au Bas-empire et à l époque des Grandes Migrations / BAR IS Oxford, Sėliai. The Selonians. Riga, Sjøvold T. The Scandinavian Relief Brooches of the Migration Period. An Attempt at a New Classification. Oslo, Svoboda B. Čechy v dobĕ stĕhováni národů. Praha, Tautavičius A. Taurapilio kunigaikščio kapas // Lietuvos Archeologija Tejral J. Zur Chronologie der frühen Völkerwanderungszeit im mittleren Donauraum // Archaeologia Austraica Tejral J. Zur Unterscheidung des vorlangobardischen und elbgermanisch-lanogbardischen Nachlasses // Die Langobarden. Herrschaft und Identität. Wien, Vallet F. A propos des tombes à épées d apparat de la Rue Saint-Pierre (Oise) et d Arcy-Sainte-Restitue (Aisne) // Revue Archéologique de Picardie Uenze S. Die spätantiken Befestungen von Sadovec (Bulgarien). München, Žulkus V. The Lower Reaches of the Nemunas (Memel) and Prieglius (Pergel) // Transformatio Mundi. The Transition from the Late Migration period to the Early Viking Age in the East Baltic. Kaunas, Амброз А. К. Фибулы юга европейской части СССР / САИ. Д М., Амброз А. К. Фибулы из раскопок Танаиса // Античные древности Подонья Приазовья. М., Амброз А. К. Южные художественные связи населения верхнего Поднепровья в VI в. // Древние славяне и их соседи / МИА М., Аун М. Археологические памятники второй половины 1го тысячелетия н. э. в юго-восточной Эстонии. Таллин, Баран В. Д. Черняхiвська культура. Київ, Веймарн Е. В., Айбабин А. И. Скалистинский могильник. Киев, Гавритухин И. О. Кодынские фибулы (типы и некоторые проблемы интерпретации) // Vakarů balt ů archeologia ir istrija. Klajpeda, Гавритухин И. О. Комплексы пражской культуры с датирующими вещами // Archeologia o początkach Slowian. Kraków, Казанский М. М. О балтах в лесной зоне России в эпоху Великого переселения народов // АВ Казанский М. М. Реки восточной части Балтийского бассейна и античные географы. Еще раз о Турунте и Хесине // АВ. 2010a. 16. Казанский М. М. Скандинавская меховая торговля и «Восточный путь» в эпоху переселения народов // Stratum Plus. 2010b. 4. Казанский М. М. О контактах населения территории восточной Эстонии и Скандинавии в эпоху Великого переселения народов // Диалог народов и культур средневековой Европы. К 60-летию со дня рождения Евгения Николаевича Носова. СПб., Кулаков В. И. «Звериноголовые» фибулы балтов (V VII вв.) // СА Кулаков В. И. Варникам. Древности пруссских вождей // Гiстарычна-Археалагiчны Зборник Кулаков В. И. История Пруссии до 1283 года. М., Кухаренко Ю. В. Могильник Брест-Тришин. М., 1980.

174 174 Michel KAZANSKI Левада М. Е. Сухоносивка // Terra barbarica. Studia ofiarowane Magdalenie Mązyńskiej. Lódź; Warsawa, Медведев А. М. К вопросу о происхождении и хронологии украшений эпохи великого переселения народов (фибулы типа Вильканцы) // Матэрыалы па археалогii Беларусi Митрофанов А. Г. Железный век средней Белоруссии (VII VI в. до н. э. VIII в. н. э.). Минск, Некрасова А. Н. Памятники черняховской культуры Днепровского Левобережья // Готы и Рим. Киев, Поболь Л. Д. Находки римских монет на землях Беларуси // Lietuvos Archeologija Погодин А. А. Из истории славянских передвижений. СПб., Петрашенко В. О., Циндровська Л. О. Слов янськi матеріали з Канівського Подніпров я // Археологія Приходнюк О. М. Пеньковская культура. Воронеж, Русанова И. П., Тимощук Б. А. Кодын славянские поселения V VIII вв. на р. Прут. М., Савельева Е. А. Олай Магнус и его «История северных народов». Л., Седин А. Предметы вооружения из городища Никодимово // Acta Archaeologia Albarutenica Скворцов К. Н. Могильник Митино V XIV вв. (Калининградская область). По результатам исследований 2008 г. I. М., Смиленко А. Т. Городище Башмачка III IV вв. н. э. Киев, Сыромятников С. И. Балтийские Готы и Гута-Сага // Живая Старина /1. Терпиловский Р. В. Славяне Поднепровья в первой половине I тысячелетия н. э. Lublin, Третьяков П. Н. Древности второй и третьей четвертей I тыс. н. э. в Верхнем и Среднем Подесенье // Раннесредневековые восточнославянские древности. Л., Федоров Г. Б. Малаештский могильник // Черняховская культура / МИА. 82. М., Фурасьев А. Г. О времени и обстоятельствах гибели городища Демидовка в верховьях Днепра // Ладога и Северная Европа. Чтения памяти Анны Мачинской. СПб., Шадыра В. I. Беларускае Падзвiнне (I тысячагоддзе н. э.). Мiнск, Шмидт Е. А. Верхнее Поднепровье и Подвинье в III VII вв. н. э. Тушемлинская культура. Смоленск, Щукин М. Б. Готский путь. СПб., Эпоха меровингов. Европа без границ. Berlin, 2007.

175 Water Routes from the Baltic to the Black Sea and Northern Barbarians in the Migration Period 175 Михаил КАЗАНСКИЙ Водные пути из Балтики к Чёрному морю и северные варвары в эпоху Переселения народов Резюме Хорошо известны три водных пути из Бал тийского бассейна к Понту, которые функ ционировали в римское время: вислен ско-бугский, неманско-березинский и двин ско-днепровский. Имеются археоло гические данные, свидетельствующие об их существовании и в эпоху Переселения на ро дов. Наиболее известна водная дорога из Южной Прибалтики по Висле и Западному Бугу, далее в бассейн Днестра и Южного Буга, хорошо намечаемая по целому ряду категорий археологического материала позднеримского времени. Для Переселения народов письменных источников о функционировании этого пути не имеется. Однако ряд находок этой эпохи на Украине и в Северном Причерноморье свидетельствует о функционировании вод ной дороги по Висле и Западному Бугу, кото рая, так или иначе, смыкалась с водными комму никациями по Днестру, Южному Бугу и бассейну Днепра. Упомянем в первую очередь находки в верховьях Днестра и Прута. Это наконечник ремня изогнутой формы, найденный на поселении Рипнев II. Такие наконечники типичны для западнобалтских культур Пруссии и германцев вельбарской культуры периода D1 по хронологии ев ропейского Барбарикума (360/ /410 гг.). Вероятно, по висленскому пути попала на верхний Прут и арбалетная фибула типа «Прага-Вильканцы», встреченная на поселении пражской культуры Кодын I. Можно предполагать, что фибула из Кодына свидетельствует о каких-то связях между юговосточной Прибалтикой и Приднестровьем, осуществлять которые логичнее всего по водным путям бассейна Вислы. Но «северные» вещи эпохи Переселения народов известны и восточнее, вплоть до Крыма и Нижнего Дона. Это подвязные фибулы с пластинчатой, иногда декорированной спинкой и с кнопкой на головке, похожие на застёжки типа «Альмгрен 163» видимо, их дериваты. Зона основного распространения таких фибул совпадает с ареалом вельбарской культуры, где они имеют довольно широкую дату. В Северном Причерноморье ко времени Переселения народов относятся находки таких фибул в погребении 421 могильника Скалистое, в некрополе Алмалык-Дере, а также в Танаисе. В эпоху Пере селения народов в зоне черняховской куль туры отмечается усиление некоторых ино культурных элементов северного проис хождения, в частности, пшеворских. Как раз в это время, в начале эпохи Переселения народов, обезлюдивают Центральная и Северная Польша, т. е. зоны пшеворской культу ры. К. Годловский полагал, что де мо графический спад к северу от Карпат связан с уходом большой массы населения на юг, под «протекцию» гуннов. Можно предположить, что какая-то часть но сителей пшеворской культуры мигрировала в черняховскую зону. Это одна из возможных причин активизации висленско-бугского пути в эпоху Переселения народов. Судя по археологическим данным, путь по водной трассе Неман Вилия Березина Днепр, существовавший в римское время, функционирует и в эпоху переселения народов. На это указывают монетные находки конца IV в. на Немане, а также появление серии западнобалтских вещей в Поднепровье. Это арбалетные фибулы с лопатковидной ножкой (Гуры, Каменево-2, Григоровка-Чернечий Лис, район Канева Переяслава, Лащова), а так же фибулы типов «Шонверлинг» (Бабина Го ра), «Сенсбург-Мронгово» (Шульговка), «Доль кайм-коврово» (Поднепровье). Есть в ра йоне неманско-березанской трассы и более экзотические вещи в верхнем слое го родища Лабенщина (тушемлинская культу ра) на одном из притоков Свислочи в бассейне Березины найден гот ландский ременной наконечник типа «Stora Hästnüs». Показательно, что для эпохи Пе реселения народов фиксируется заметное при сут ствие скандинавских элементов в начале рас-

176 176 Michel KAZANSKI сматриваемой здесь трассы от южной части Куршского побережья до древнего залива к северу от современного города Эльблонг. Функционирование неманско-березанского пу ти, как бы в обход бассейнов Вислы, Одера и Ду ная, возможно, связано с какими-то конкретными политическими событиями в Центральной Европе, где время от времени, например при вторжении гуннов и сразу после смерти Аттилы, вспыхивали жесткие вооружённые конфликты, временно нарушавшие сложившуюся систему трансъевропейских связей. Еще один возможный путь проникновения северных варваров в лесную зону Восточной Европы проходил по Рижскому заливу и Западной Двине. В устье Западной Двины, пустующем в римское время, в V в. на левом берегу появляется население, оставившее мо гильник Катлакална Плявниекалнс с грунтовыми могилами. Это может быть связано с активизацией речных и морских трасс в районе устьев рек Западная Двина и Гауя в эпоху Великого переселения народов. Возможно на существование в конце эпохи Переселения народов какой-то трассы вдоль южного берега Рижского залива, так или иначе связанной с устьями Западной Двины и Гауи, указывают находки ранневизантийских монет. Находки фибул западнобалтского происхождения на нижнем участке течения Западной Двины также свидетельствуют о том, что какие-то контакты по этому водному пути осуществлялись и в эпоху Переселения народов. Клад из Браслава, содержавший золотую монету Валента ( гг.), найденный в Подвинье, недалеко от современной латвийско-белорусской границы, как бы маркирует конечную восточную точку западнодвинской речной магистрали. Однако вверх по Двине, в зону тушемлинской культуры, западнобалтский импорт не идёт. Для эпохи Переселения народов в ареале тушемлинской культуры фиксируется серия «южных» импортов, со Среднего Дуная и из южной части Восточной Европы. Эти находки из понто-дунайской зоны могут свидетельствовать о проникновении на Верхний Днепр в конце IV первой середине V в. групп «южного» и «западного» населения. Разрушения военного характера, относящиеся к этому времени, отмечены на городище Демидовка. Похоже, что эти события и дестабилизировали днепро-двинский путь.

177 177 Igor KHRAPUNOV Germanic Artefacts in the Cemetery of Neyzats Various categories of artefacts of Germanic origin discovered in the Crimea have been subjects of scholarly research many times. As a rule, they have been studied within the context of the chronology and circumstances of the Germanic tribes, and primarily the Goths, migration to the Crimea and their settling in the peninsula. Erast Symonovich has discovered in Crimean museums collections grey-clay wheel-made vessels similar to the Chernyakhov ware in his point of view. He has supposed that these artefacts appeared in the Crimea in result of trade transactions (Сымонович 1975). Vladislav Kropotkin has listed groups of artefacts found in the Crimea with direct analogies in the Chernyakhov area that appeared in the peninsula in result of a migration of the tribes of Gothic union. His list includes various types of ceramic artefacts, radiate-headed brooches, fibulae with cur ed bow and returned foot, bone combs and pyramidal pendants, metal bucket-shaped pendants and pendants in the form of Thor s hammer, temple rings with beads, amber mushroom-shaped pendants, carnelian 14-hedral beads, shield bosses and grips, and gold granulated lunula pendants (Кропоткин 1978). Igor Pioro has supplied Kropotkin s corpus with new finds and new categories of finds (battle axes, ritually bent sword) that were brought to the Crimea from the Chernyakhov area. He has stated poor development of the commodity production and, therefore, inter-tribal trade among the Chernyakhov tribes and has used it as a background to the conclusion that the people of that culture penetrated into the Crimea (Пиоро 1990: ). Kropotkin and Pioro have mentioned that the artefacts in their lists came to the Crimea from the Chernyakhov area. It is not probably correct in regard to shield parts, because weapons are almost not known in the Chernyakhov cemeteries. More likely, the Crimean shields originate from the Przeworsk area. Michel Kazanski has made a large contribution to the list of the Germanic artefacts in the Crimea and has widened its nomenclature with strap-end Raddatz J2-1, Т-shaped brooches, and amber double-axe-shaped pendants. In regard to the culture and chronology, he has divided the finds from the period C2 D1 of our present interest into two stages. The finds from the middle and second half of the third century appeared in the Crimea together with the Wielbark and possibly Przeworsk population and the fourth century artefacts with the Chernyakhov culture people (Kazanski 2002: ; Казанский 2006: 28 29). Kazanski is probably not quite correct when adding red-enamelled spurs from the cemetery of Skalistoye III to his list: they obviously originate from the Balts land or the Middle Dnieper area. Kazanski has provided a particular analysis of the Germanic artefacts discovered in the Bosporan kingdom area. He has run to the conclusion that all them belong to the Chernyakhov culture and all appeared in Bosporos in the fourth century AD, i. e. later than in the Crimean foothills (Казанский 1999: 280). Vladislav Yurochkin has compiled a corpus of 34 grey-clay vessels from the south-western Crimea. In his opinion, this ware appeared in the peninsula with the tribes of the Chernyakhov culture (Юрочкин 1999). This brief review makes clear that some artefacts from the areas of the cultures shaped under the Germanic influence, and first of all the Chernyakhov culture, circulated in the Crimea in the third and fourth century. The excavations of the cemetery of Neyzats allow one to enlarge and amend the list of the Germanic artefacts in the Crimea. The cemetery of Neyzats is located in the centre of Crimean foothill area, about 20 km east of Simferopol. During the period of excavations that started in 1996, we investigated 518 graves, including burial vaults, undercut and simple pit graves. The cemetery dates from the second to fourth century AD. I have published general review of the

178 178 Igor KHRAPUNOV results of the cemetery excavations and all categories of grave goods uncovered (Храпунов 2011a). Among the abundant grave goods, there is a small group of artefacts that could be related to the appearance of the Germanic tribes in the Crimean peninsula with a greater or lesser degree of probability. Shield boss (fig. 1. 1). This iron cylindricalconical artefact was discovered in vault no. 306 with numerous materials from the third and fourth century AD. Its dimensions: 15.3 cm high, 13.4 cm in diameter; collar 1.5 cm wide; spike 7.0 cm long. This shield boss belongs to Martin Jahn s type 7a (Jahn 1916: , Taf. III), type I of the Gutteberg group of Germanic weapons in Norway (Bemmann 1994: 183; Bemmann, Hahne 1994: , 458, 460), and type 3b of shield bosses discovered in Illerup bog in Denmark (Ilkjaer 1990: 35, 330). Kazimierz Godłowski has attributed similar shield bosses to groups 4 and 5 of the Przeworsk weapon graves (Godłowski 1992: 80, 82, ryc. 2. 1; 3. 1, 2). The listed shield bosses were widespread in the second and third century AD (phases B2b to C1a of the European timeline). Many scholars consider that shields with bosses were brought to the northern Black Sea area, the Crimea in particular, with the Germanic participants of Gothic raids (Ščukin 1993: 326; Каргапольцев, Бажан 1992: 118). All the researchers consider that shield bosses were absolutely not typical of the Sarmatian and Scythian weaponry and appeared in the northern Black Sea area together with the Germans. Indeed, shield bosses are extremely rare in the Sarmatian graves so their appearance is always, and quite logically, related to borrowing from the outside. The Late Scythians never had shields with bosses at all. All the cultures shaped with Germanic participation have a great number of shields with metal parts. However, it is interesting to note that shield bosses of the type discovered in Neyzats are not known in the Chernyakhov culture which is the closest to the Crimea, as well as in the Sarmatian and Crimean sites (for the list of finds see: Kazanski 1994: ). Their closest geographically and very exact analogies appear in the Przeworsk area. Most likely, the shield boss from Neyzats appeared in the Crimea before the Germanic raids recorded by written sources (Храпунов 2010a: ; 2011b: 8). Shield grip (fig. 2. 1) of iron. Fragmented artefact does not allow the reconstruction of its length, other dimensions are 19.0 x 2.5 cm. Discovered with the just described shield boss. The shield grip cannot be attributed to a certain type due to its poor preservation. It is most similar shield grips of type IV from Norway (Bemmann, Hahne 1994: , Abb. 95) and probably corresponds to Jorgen Ilkjaer s type 5c (Ilkjaer 1990: 36, Abb ). Shield grips of type 5c in Scandinavia date back to the second half of the third or fourth century AD (phases C2 C3). A simi lar handle was discovered in grave no. 3 in the cemetery of Chatyr-Dag from the second half of the third to the early fourth century (Мыц et al. 2006: ). Neither in Scandinavia nor on the side of Chatyr-Dag mountain we have grips 5c accompanied with boss 3b, as it was in vault no. 306 of the cemetery of Neyzats. However, the grip from Neyzats is ill-preserved, so its type could be misattributed (Храпунов 2010a: ; 2011b: 8 9). Shield boss of bronze, with cylindrical calotte terminating with spike (fig. 1. 2) was discovered in grave no. 152 with the materials from ca AD. Dimensions: 9.5 cm high, 15 cm in diameter, collar 2.5 cm wide. There were four symmetrical holes in the collar with bronze rivets inserted into three of them. Top spike is missing. This shield boss was repaired in antiquity. A bronze plate was inserted into the pin, with its top bent over the end of the broken pin and fixed with a ring of bronze wire. There were fragments of bronze clip discovered amidst disturbed bones; they assembled into a ring 15 cm in diameter that exactly corresponds to the diameter of the shield boss. This clip possibly fit to the edge of the collar. The defect of the shield boss in question (a great part of its spike is missing) complicates its attribution according to existing classifications. It is similar to Martin Jahn s types 7a and 7b (Jahn 1916: , Taf. III), Norbert Zieling s type D (Zieling 1989: 71 73, Taf. 7), types I and II in the Gutteberg group and Vennolum group of Scandinavian weapon graves (Bemmann, Hahne 1994: ), Ilkjaer s type 3 (Ilkjaer 1990: 35, 330), and various artefacts from Godłowski s groups 3 5 of the Przeworsk weapon graves (Godłowski 1992: 72, 80 82). The mentioned types of shield bosses existed in the second and early third century AD (phases B2a C1a of the European timeline). The latest finds comparable with the Neyzats one (Konin type) date back to AD From the second half of the third century AD onwards, there mainly were hemispherical and conical shield boss-

179 Germanic Artefacts in the Cemetery of Neyzats 179 Fig. 1. Shield bosses: 1 grave no. 306, iron; 2 grave no. 152, bronze es, without a spike, so they were definitely other than the artefact under study. The individual characteristics of the Neyzats find resemble shield boss from Thracian tomb in Karaagach more than other artefacts (Велков : 24, обр. 24), as well as shield boss from Hauran in southern Syria. The Karaagach shield boss dates back to phase B2 (AD /180), though the find from Hauran belongs to Kazanski s type Vermand (Kazanski 1994: 436, ). However, Vermand shield bosses do not have spikes. Shield bosses like that of Neyzats are not found in the Crimea (for a list of Crimean finds see: Kazanski 1994: ). In geographical terms, the most close to the Neyzats find is the shield boss of type Konin discovered in subterranean room covered with soil during the Gothic devastation of Tanais in the mid-third century AD (Арсеньева, Шелов 1974: 135, табл. XV. 6). Despite of the similarity between the shield boss from grave no. 152 and many Przeworsk finds, all the Przeworsk artefacts in contrast to the one from Neyzats are made of iron. Therefore we still have doubts concerning its Przeworsk or generally Germanic origin (Храпунов 2003: ; 2010a: 551; Khrapunov 2005: ). Axe of iron with straight top edge, oval hole, and sickle-shaped blade (fig. 2. 2) was discovered

180 180 Igor KHRAPUNOV Fig iron shield handle, grave no. 306; 2 iron axe, grave no. 4; 3 5 artefacts entwined of metal rings (3 grave no. 59, 4 5 grave no. 208) in vault no. 4 with numerous materials from the fourth century AD. The other axes from the cemetery of Neyzats, similarly to almost every axes from the Late Roman period discovered in the Crimean sites, have pronounced butt to differ them from the above-mentioned find. The axe in question without pronounced butt belongs to the same type as almost all the axes in the Chernyakhov culture (Kokowski 1993: 337; Магомедов, Левада 1996: 308). The Neyzats axe belongs to Grzegorz Kieferling s western series of group 5. Axes of this series from the Roman and early Great Migration periods were discovered in the Chernyakhov, Wielbark, and Przeworsk cultures, as well as in the Elbe area, in the south-west Germany, and in Bohemia (Kieferling 1994: 343). The find of this axe, in contrast to other Neyzats and other Crimean finds, suggests Sarmatian-Germanic contacts (Храпунов 2008: 370; 2010a: ). Swords (fig ). In the cemetery of Neyzats, there were more than ten long swords without metal pommel and cross-guard (Храпунов 2010a: ). Aleksandr Simonenko has noted that long swords from the Crimean sites from the second half of the third and fourth century AD belong to Anatoliy Khazanov s type 3, with the blade at a right angle to the pin. This feature is similar to

181 Germanic Artefacts in the Cemetery of Neyzats 181 Germanic swords, 1 thus making them a symbiosis of Romano-Gothic and Sarmatian weapons (Симоненко 2010: 58). I think that this solution is too complicated for the question. Sarmatian world knew type 3 of swords very well (Хазанов 1971: 17; Безуглов 2000: 171). Almost all the weapons in the cemetery of Neyzats are Sarmatian, hence, the origin of the long swords is also Sarmatian to follow Occam s razor. They all originate from the fourth century assemblages and have features typical of the swords from this period: wide blades, relatively short tangs of handle, the lack of pommel (Безуглов 2000: 181). The feature correctly understood by Simonenko that almost all the Crimean swords including those discovered in the cemetery of Neyzats are of Khazanov s type 3 could be explained by their local origin. Crimean blacksmiths possibly developed a tradition of making swords with the grip pin meeting the blade at a right angle. Artefacts entwined of metal rings (fig ). Excavations of several undercut graves in the cemetery of Neyzats discovered artefacts of iron rings interwoven like a chain mail. They accompanied burials of women from the mid-second to the midthird century AD. Similar goods were found in Crimean cemeteries of Bel bek IV, Skalistoye III, Chyornaya River, Zavetnoye, Bitak, and Ust - Al ma. In all reliable cases, the artefacts of entwined iron rings accompanied women s graves. As a rule, these burials are made into undercut graves from the second to the first half of the third century AD. Location of these iron artefacts on skel- Fig. 3. Iron swords, grave no Literally, weapons from Gothic sites of the Przeworsk and Wielbark cultures, but it is a kind of confusion: the Przeworsk culture was not created by the Goths, though the Wielbark graves did not contain weapons.

182 182 Igor KHRAPUNOV etons was not stable. The number of artefacts is one to four pieces in a grave. They are 2 to 7 cm long (Пуздровский 2007: 138; Бабенчиков 1963: 108; Богданова, Гущина, Лобода 1976: 146, рис ; Ахмедов, Гущина, Журавлёв 2001: 183, рис ; Зайцев et al. 2007: 259, рис ; Храпунов 1999: 263, рис. 4. 8; Храпунов 2007: 34, 42 45, рис. 9. 9). All the researchers who excavated the artefacts in question suppose that they are dealing with chain mail fragments. However, this attribution leaves space for doubts, because the artefacts of iron rings are discovered only in graves of women and are of small size. Obviously, it is not the case of small remains of bigger artefacts, but actually small things put into graves. The finds analogous to the Crimean were made in some vaults in the Golden Cemetery in the Kuban area. There were small artefacts of interwoven iron rings sometimes in the same graves with plates and scales of armours. It was the background for the conclusion of composite armour to exist in the Kuban area in the Roman period. However, the vaults are plundered, and the necessary field documents are absent, so we have enough possibilities for various hypotheses. Location of the artefacts of interest can not be determined, and they could not be related to this or that burial (Хазанов 1971: 60 62; Гущина, Засецкая 1994: 10 11). There are more, and I think more close, analogies in the Przeworsk culture area. The finds of interwoven iron rings, analogous to the Crimean ware, have been discovered in many women s graves from horizon B2/C2; in other words, they date back to ca AD. A hypothesis states that the Germanic warriors during the Marcomannic Wars took mails off killed Romans, and their women used these fragments as amulet ornaments with traditional protective function (Kontny 2004: 155). Analogous finds were discovered in the area of the Chernyakhov culture as well, but they are rare and later than the Crimean and Przeworsk artefacts (Тиліщак 2010; 2011). The similarity of artefacts, their specificity, synchronism, as well as relation to female graves, despite of remoteness of the territories, allows me to suppose contacts between the populations of the Crimean foothills and modern Poland territory. Taking the absence of similar finds in intermediate territories into account, such contacts could be called immediate rather than indirect. The function of the artefacts of intertwined iron rings remains unclear. The only understandable thing is that they were inalienable element of the material culture, or, precisely, of female subculture of the population of Crimean foothills in the Roman period (Храпунов 2010a: ; Храпунов 2010b). Grey clay wheel-made vessels. The excavations of the cemetery of Neyzats have uncovered more than 1,000 hand-made and about 500 red slip vessels. A small group of wheel-made vessels with grey surface stands against this background. There is a possibility to divide the finds of Neyzats according to Boris Magomedov s classification of the Chernyakhov vessels (Магомедов 2001: 47 48) and Vladislav Yurochkin s classification of grey-clay vessels from the south-western Crimea (Юрочкин 1999: ). It is necessary to note that Magomedov s classification of the Chernyakhov bowls considers close or open form of the vessel as well as the presence and the location of cordons on its shoulders, but it does not take into account the form of vessels, so one type could unite bowls of considerably different morphology. Five vessels have been found in vault no. 281 which is among the latest graves in the cemetery. It dates back to the second half or even the end of the fourth century AD. Vessel (fig. 4. 5) corresponds to Magomedov s type 2в of bowls. Similar ornamentation rarely appears on the Chernyakhov pottery, when cordons are sometimes divided and edges of the body are sometimes cut (Гудкова 1986: 152, 157, рис ; 5. 1, 8; Сымонович 1964: 320, рис ). Vessel (fig. 4. 6) corresponds to type 1 (with plain shoulder) of the Chernyakhov bowls. Vessel (fig. 4. 9) corresponds to Magomedov s type 2в and Yurochkin s type 5. Vessel (fig. 4. 7) corresponds to Magomedov s type 2в. Vessel (fig. 4. 8) corresponds to Magomedov s type 2в and Yurochkin s type 6. Vessel (fig. 4. 4) corresponds to Magomedov s type 2a and Yurochkin s type 6. Two more vessels were uncovered in the fourth century vault no Vessel (fig ) corresponds to Magomedov s type 2в and Yurochkin s type 6. Vessel (fig. 4. 2) belongs to open forms of bowls in the Chernyakhov culture. The ornamentation of grey-clay vessels in Neyzats corresponds to the Chernyakhv culture. There are zigzags and triangles (fig. 4. 4, 8, 10), drawn by a wheel or combed stamp to form a hori-

183 Germanic Artefacts in the Cemetery of Neyzats 183 Fig. 4. Grey-clay vessels: 1 grave no. 461; 2, 10 grave no. 305; 3 grave no. 476; 4 9 grave no. 281; 11 grave no. 4 zontal frieze on the vessel s shoulder (Магомедов 2001: 57). Almost all the grey clay wheel-made vessels in Neyzats find direct analogies in the Chernyakhov culture. They compose a small group, certainly different from the most part of the vessels discovered in the cemetery. The differences are the type of surface treatment, clay colour, and bi-conical form of the body: the latter feature is extremely typical of the Chernyakhov pottery and absolutely not to red slip and hand-made ware in the cemetery of Neyzats. This is the reason to suggest the Chernyakhov origin of the grey clay pottery in Neyzats and other grey-clay ware from the cemeteries of Crimean barbarians. However, there is one thing to raise doubts. Excavations of the cemetery of Neyzats have found wheel-made grey clay bowls with black coating in the fourth century AD vaults nos. 4, 461, and 476 (fig. 4. 1, 3). Similar vessels have been unearthed in the cemeteries of Suvorovo (Юрочкин, Труфанов 2003: 202, рис. 29), Krasnaya Zarya (Неневоля, Волошинов 2001: 143, рис. 5. 2), Tas-Tepe (Пуздровский, Зайцев, Неневоля 2001: рис ), and Inkerman (Веймарн 1963: 39, рис. 13, 14, 15). We can

184 184 Igor KHRAPUNOV Fig. 5. Glass vessels: 1 grave no. 139; 2 grave no. 301; 3 accumulation of vessels no. 15; 4 6 grave no. 275 call this type one of the most widespread among wheel-made grey clay ware in the Crimea. There is nothing similar in the Chernyakhov culture. Therefore, we can infer the existence of some non-chernyakhov centre where grey clay wheelmade pottery was produced, whence it later penetrated to Crimean barbarians. Glass vessels. The excavations of Neyzats uncovered more than 150 glass vessels so far. All of them are published (Шабанов 2011). Some of the glass vessels were brought to the Crimea from the area of Germanic cultures. All the vessels described below are uncovered from the fourth century assemblages. This way, the vessel of Eggers type 223 from grave no. 139 (fig. 5. 1) has only two analogies, one in the Chernyakhov area (Петраускас, Пастернак 2003: 68) and another in Danish Zealand island (Eggers 1951: 180). Graves nos. 275 and 301, and also the accumulation of vessels no. 15 contained five beakers of Eggers type 230 (fig ). These vessels were distributed very wide, from the northern China to Scandinavia, though their maximum concentration was recorded in the Chernyakhov culture (Eggers 1951: 180, Karte 58; Rau 1972: 124, ; Straume 1987: 29, Karte 2; Pánczál, Dobos 2007: 69), whence they probably came to the Crimea. Beaker from pit with vessels no. 6 (fig. 6. 1) and two conical beakers from vaults nos. 115 and 485 (fig ) are decorated with drops of blue glass. Such vessels are discovered in many provinces of the Roman empire and outside its borders. We should underline their exclusive rarity in Crimean cemeteries like Neyzats. Only two conical beakers more are discovered in the cemetery of Suvorovo (Зайцев, Мордвинцева 2003: 58, 61, рис ;

185 Germanic Artefacts in the Cemetery of Neyzats 185 Fig. 6. Glass vessels: 1 pit with vessels no. 6; 2 grave no. 115; 3, 5 grave no. 485; 4 grave no. 424; 6 accumulation of vessels XI ). In the Chernyakhov area, conical beakers appeared probably later, than in other places, in the late fourth and early fifth century (Гавритухин 2000: ). This chronology coincides with the cease of the cemetery of Neyzats and similar sites in the Crimea. If one agrees that the blue drops beakers came to the Crimea from the Chernyakhov area, it could happen only in the last years of that culture, which is why the number of the beakers in Crimean graves is small. Conical vessels with cast base and applied thread decoration have been discovered in vault no. 485 (fig. 6. 5) and in pit with vessels no. 11 (fig. 6. 6). Although exact analogies to these vessels are not known, there are few artefacts of similar form and ornamentation in the Chernyakhov area (Гудкова, Фокеев 1984: 83, рис ; Кропоткин 1970: 30, рис ; Россохацкий 1987: 147, рис. 1. 4), as well as in Norway (Eggers 1951: 178, Taf ) and in a glass-making workshop in Cologne (Doppelfeld 1966: fig. 132). It should be underlined again that all the vessels indicating the north-western connection of the population who created the cemetery of Neyzats originate from the fourth century assemblages and no earlier piece is discovered. Combs. The excavations of the cemetery discovered nine more or less fragmented bone combs. There is a classification of this category of artefacts by Sigrid Thomas, based on the materials of mostly Central and North Europe, and much detailed classification of the finds in the Chernyakhov culture area. Two combs from graves nos. 156 and 492 are of type Thomas I (fig ). These combs were produced from the second half of the second to the late fourth century AD and were especially popular in the second half of the third and first half of the fourth century AD (Thomas 1960: 77 94). They belong to variant IB1 of the Chernyakhov classifi-

186 186 Igor KHRAPUNOV Fig. 7. Bone combs: 1 grave no. 156; 2 grave no. 492; 3 grave no. 312; 4 grave no. 281; 5 grave no. 12; 6 grave no. 19 cation. Combs of this variant are discovered in the fourth century AD burials (Магомедов 2001: 84). Combs from graves nos. 12, 19, 281, and 312 belong to type Thomas III (fig ), dated by the researcher to the second half of the fourth and fifth century AD (Thomas 1960: ), or the latest variant IIIВ1а of the Chernyakhov combs from the last quarter of the fourth and first decades of the fifth century AD. The maximum concentration of combs of this type are in the Chernyakhov culture area (Магомедов 2001: 84). Fragments of combs from graves no. 275 (two specimens) and 306 are too small to understand the type of complete pieces. The researchers unanimously consider that bone combs came to the Crimea from the area of East Germanic cultures (Кропоткин 1978: 153; Пиоро 1990: 101, 103; Казанский 1999: 277). Horn bracelet. It was discovered in the fourth century AD vault no The bracelet measuring 7.5 х 6.1 cm, remained as fragments (fig. 8. 4). Such goods are sometimes discovered in the area of Germanic cultures Przeworsk, Masłomęcz group of the Wielbark, Sântana de Mureş, and Chernyakhov. In a few cases, they were, as their small size and position in graves suggest, amulet pendants rather than bracelets (Kokowski 2004: 28 30, 45; Магомедов 2001: 74). The bracelet from the cemetery of Neyzats is the only one discovered in the Crimea so far. There is no doubt that it was brought to the peninsula from the area of Germanic cultures. Fibulae and brooches. Bronze radiate-headed brooch 4.1 cm long was discovered in one of the latest in this cemetery graves no. 73 (fig. 8. 1). It belongs to Anatoliy Ambroz s variant IББ (Амброз 1966: 83). A few finds of such clasps, the earliest among the radiate-headed, are rather evenly distributed among the entire Chernyakhov area, though small radiate-headed brooches are very few outside of it (Амброз 1966: 83, 86; Kokowski 1996: ; Gavritukhin 2002: ). According to the distribution area of small radiate-headed brooches, they penetrated to the Crimea from the Chernyakhov area. Such artefacts are ab-

187 Germanic Artefacts in the Cemetery of Neyzats 187 Fig bronze brooch, grave no. 73; 2 fragment of a bronze fibula, grave no. 57; 3 iron fibula, grave no. 4; 4 worked horn, grave no. 474; 5 6 gold ornaments, grave no. 169 sent in all synchronous cemeteries of the same culture in the Crimea. The probable explanation is that the clasp of Neyzats was brought to the Crimea in the final stage of this cemetery and other cemeteries of the same culture (Храпунов 2005: ). There is a fragment of two-piece bronze fibula with curved bow, a hole for spring axle, and a cord-hook (fig. 8. 2). Complete specimens of such brooches have comb on the bow and high catch plate. This fragment is discovered in grave no. 57. It was used with beads as a pendant in a necklace (Khrapunov 2001: 271, fig. 1. 5; Храпунов 2006: 174, рис ). This fragment is of fibula A83 or A84. We cannot determine it with more precision because many details important for classification are missing

188 188 Igor KHRAPUNOV Fig. 9. Pendants: 1 grave no. 178, gold; 2 grave no. 169, iron; 3 4 grave no. 208, iron; 5 grave no. 29, iron; 6 11, 18 grave no. 4, amber; 12 grave no. 325, amber; 13 grave no. 473, bronze; 14 grave no. 193, amber; grave no. 107, amber (Almgren 1973: 42 43). The main concentration of such fibulae was in the northern part of Roman province Pannonia. Whence they spread in all directions, with their number getting smaller the further they came. The easternmost find places of fibulae А84 are in the Lower Dnieper area and in Olbia (Almgren 1973: ; Амброз 1966: 38; Dąbrowska : 14, ryc. 6). Fibulae of the type date from the second and the early third century AD (Dąbrowska : 18 19). So far the fragment from the cemetery of Neyzats is the easternmost find of fibulae А Undoubtedly, the Neyzats find reflects some western connection of the Sarmatian population in the Crimea. We cannot be sure about the region whence this fibula came to the Crimea, was it, for example, Pannonia or the Przeworsk culture area, where a number of such clasps was discovered. The excavations of the cemetery have discovered a number of iron two-piece fibulae with narrow foot and low cord, of the so-called warrior type (fig. 8. 3). This is specific Crimean type of clasps. Taking into account their construction similarity to fibulae of Ambroz s group 17, sub-group 1 (Амброз 1966: 70 71), I have supposed that they were produced according to the bronze samples from the Chernyakhov culture (Храпунов 2002: 55). Maxim Levada has paid attention to fragments of two ornaments from grave no They are made of thin embossed gold plate and connected to bronze backing that was fastened to a leather strap (fig ). He has supposed that these are parts of brooches, but it is most likely not true. Anyway, such an ornament finds direct analogies in Scandinavia. All the researchers dealing with the Germanic penetration into the Crimea have paid attention to fibulae with curved bow and returned foot, which are sometimes called Chernyakhov brooches. They are numerous in the Chernyakhov area and in the Crimea, so Neyzats is not an exception in this regard (Храпунов 2011a: 28). In both regions they had simultaneous and similar development. This often is a background for the conclusion of the contacts between the Chernyakhov population and the residents of the Crimean foothills. However, we have no limits to prevent the suggestion of the opposite direction of these contacts or to imagine some centre that produced these clasps and distributed them to both regions. The same observation is valid for the buckles, similar in the Crimean and Chernyakhov cemeteries. Bucket-shaped pendants were made of gold (fig. 9. 1), silver (fig. 9. 2), and iron (fig ). They are discovered in graves no. 29, 169, 178, and

189 Germanic Artefacts in the Cemetery of Neyzats 203. Some of the Neyzats pendants have a hole in the bottom, so they were probably used as beads. Pendants of the type are listed in all the aforesaid reviews of the Germanic antiquities in the Crimea. Igor Bazhan and Sergey Kargopol tsev have produced a rather logical hypothesis concerning the distribution of bucket-shaped pendants. In their opinion, pendants of the type appeared no late than in the first century BC in the Greek cities of the northern Black Sea area, and thence were borrowed by the Sarmatians. The latter brought them to Central Europe, were borrowed by the Germanic population, and the latter returned them back to the northern black Sea area. The keystone of this hypothesis is the lack of the bucket-shaped pendants in the northern Black Sea assemblages from the late second and first half of the third century AD and, therefore, the impossibility to consider the northern Black Sea tradition of the making of the bucket-shaped pendants as continuous one (Бажан, Каргапольцев 1989). However, today this chronological gap appears to be filled. Almost all the bucket-shaped pendants from the cemetery of Neyzats date back to the late second and first half of the third century AD (Стоянова 2004: 297; Храпунов 2011a: 39; Беспалый 1990: 213). Now we are confident that the pendants in the form of small buckets were continuously produced in the northern Black Sea area from the first century BC to the fourth century AD. Therefore, it is either the case of two independent traditions, Germanic and northern Black Sea, of making similar artefacts, or the Germanic population borrowed this tradition from the northern Black Sea area, possibly via the Sarmatians or the Northern Thracians where the bucket-shaped pendants were also found. In both cases, Germanic and northern Black Sea tribes produced the pendants in parallel to each other. One cannot be sure if the pendants discovered in the cemetery of Neyzats were locally made or brought to the Crimea from the area populated by the Germanics. There is another Bazhan s and Kargapol tsev s argument that raises doubts. They suppose that bucket-shaped pendants first appeared in Greco-Roman cities, though in such centres as Chersonesos and Tanais their finds are literally very few. These pendants were used by farming population of Bosporos that was ethnically mixed. Their number is especially big, from the first century BC onwards, in the Late Scythian, less in Sarmatian, sites. Therefore we have every reason to infer that the idea of bucket-shaped pendants and their use and manufacture appeared among the northern 189 Black Sea barbarians and not amidst the Greeks. The finds from Ocniţa (Buridava) in Romania, the cemetery of Dolinyany, and in Magdalensberg in southern Germany (Бажан, Каргопольцев 1989: 164; Droberjar 2011: 25) are almost synchronous to the earliest Black Sea finds. Therefore, however the number of the bucket-shaped pendants discovered in the northern black Sea is considerably bigger, their convergent appearance in different centres is also possible. Axe-shaped pendant of bronze was discovered in vault no Unfortunately, this pendant disintegrated after its extraction from the grave. The drawing is by its photograph (fig ). It belongs to Anastasiya Stoyanova s type 3 of metal axeshaped pendants in the Crimea (Стоянова 2005a: 50 53). Pendants of the form were very popular in the European Barbaricum (Kokowski 1998). They initially appeared among the northern Thracians, and afterwards, via mediation of the Sarmatians in the Great Hungarian Plain where the main concentration of such artefacts was recorded, they penetrated to the Germanic tribes (Стоянова 2005a: 51 52). Because of the fourth century AD chronology of the pendant from Neyzats, it could be brought to the Crimea by the Germanic population. However, other interpretations are also possible. In the Crimea, there are few pendants of type 3 in the form of wedge-shaped axe discovered in the sites from the first to the third century AD. In other words, they had appeared in the peninsula before the Germanic population found its way to it according to written sources. Nevertheless, the number of such ornaments became considerably larger in the second half of the third and fourth century. Moreover, there is another fact to draw our attention: the number of axe-shaped pendants is much smaller in the Chernyakhov area, the closest to the Crimean Germanic habitat, than in the peninsula (Гопкало 2008: 63). Let us also remind the tradition of making axe-shaped pendants of various types and of different materials, e. g. glass, that existed in the Crimea since old times and never interrupted. Thus, it is very much possible that the Neyzats pendant was locally made. Therefore, the case of axe-shaped pendants looks like the case of bucketshaped pendants: with equal degree of probability, they could be of Germanic or local origin. It is also worth mentioning glass axe-shaped pendant from grave no. 325, which dates back to the late second or the first half of the third century AD (fig ). Glass axe-shaped pendants

190 190 Igor KHRAPUNOV of different colours and morphological features are rarely found by excavations in Greco-Roman cities (Алексеева 1978: 74; Зубарь, Мещеряков 1983: 103) and barbarian cemeteries in the Crimea (Храпунов 2002: 47). They are seemingly not discovered outside the Black Sea area. One should suppose that these ornaments were made in a small number in Greco-Roman cities and towns. Amber pendants. These pendants in the shape of a figure of eight (fig , 14 18) were discovered in many graves from the fourth century AD, mostly of children (Стоянова 2004: ). The maximum concentration of such pendants is in the basins of the Vistula, Oder, and Elbe whence they spread in every direction (Tempelmann- Mączyńska 1985: , Taf. 70). According to a popular interpretation, the figure-of-eightshaped pendants penetrated to the Crimea from the Chernyakhov area (Кропоткин 1978: ). Because the fashion of amber ornaments spread far away from the Germanic area by the fourth century AD, the above interpretation is not the only possibility (Мастыкова 1999: ; 2009: 80), though it still remains the most probable. The figure-of-eight-shaped pendants were widely distributed in the Germanic territories in the second half of the third century AD (Tempelmann-Mączyńska 1985: 83 86). In the Crimea, they are not discovered within of assemblages earlier than the fourth century AD so far (Храпунов 2003: 344). It is interesting to note that, according to experts calculations, the number of amber figure-of-eight-shaped pendants in the Crimea is three times as large as in the Chernyakhov area (Гопкало 2008: 71). There are original amber pendant decorated with grooves from the fourth century vault no. 4 (Стоянова 2004: 292, табл. VI ; Храпунов 2008: 384, рис , 31). One of them finds the only analogy in the cemetery of Nienburg on the left bank of the Elbe. It dates back to the second half of the third century AD (Стоянова 2004: 293, рис. VI. 21; Tempelmann-Mączyńska 1985: 86, Abb. 9). Grave no. 193 contained amber pendant with specific circular cross-section and small top part to separate it from common figure-of-eight pieces (fig ). It dates back to the late second or first half of the third century AD (Стоянова 2005b: 159, 161, рис ; Храпунов 2006: 187). The same ornaments are discovered in vault no. 720 of the cemetery of Ust -Al ma, which dates back to the first century AD (Puzdrovski, Zaicev 2004: 232, 234, Abb. 3. 1). They were obviously not related to Germanic tradition, but were produced in the Crimea of imported raw materials. The production of such artefacts having no analogy outside the Crimea started in the first century AD, according to the Ust -Al ma find, and continued to the Late Roman period, as it came from the above-described ornament from Neyzats and another one from Chyornaya River cemetery (Бабенчиков 1963: 92, табл. III. 17). *** The excavation of the cemetery of Neyzats discovered a certain number of artefacts of the types that, as many researchers think, were brought to the Crimea from the area of Germanic cultures. The Chernyakhov culture, as the closest to the Crimea culture of Germanic appearance, is usually interpreted as a source of the Germanic-like goods. The analysis of the finds has uncovered that the bucket-shaped pendants, axe-shaped pendants, amber pendants in the form of figure-of-eight or some part of them could be brought to the Crimea by Germanic tribes, though they also could be produced locally. Glass vessels and radiate-headed brooch were most likely transported to the Crimean foothill area from the Chernyakhov area. The persons of the Chernyakhov culture who appeared in the Crimea are most probable authors of bone combs and horn bracelet. Warrior fibulae were likely made in the peninsula probably following Chernyakhov samples. Grey-clay wheel-made vessels are probably of the Chernyakhov origin, but the fact that one of the most popular in the Crimea types does not meet with analogies in the Chernyakhov culture allows the suggestion that there was another centre of pottery-making supplying the Crimea with its products. The axe and shield with metal parts could by war booty, and the shield is of the Przeworsk rather than Chernyakhov origin. The connection with the Przeworsk culture in a very early period, established before the midthird century, is indicated by fragments of artefacts entwined of iron rings and possibly the fragment of fibula А The most important problem standing in front of the researchers of Germanic antiquities in the Crimea is how the artefacts from the area of Germanic cultures came to the Crimean peninsula. The exploration of the cemetery of Neyzats has not answered this question. There is no grave made according to a rite used by any of Germanic tribes and no grave where Germanic goods concentrated. The latter are scattered as individual inclusions

191 Germanic Artefacts in the Cemetery of Neyzats through an extraordinary number of Sarmatian grave goods. Therefore, we have no reason to infer that there were Germanic persons who buried among the Sarmatian creators of the cemetery of Neyzats. One can only speculate about the way in which Germanic artefacts penetrated into the 191 Crimea was this a result of military, trade, or cultural contacts, and where exactly these contacts happened. All the above speculations are possible theoretically, though none of them could be proved by the analysis of this or that assemblage excavated in the cemetery of Neyzats. Bibliography Almgren O. Studien über nordeuropäische Fibelformen. Bonn, Bemmann J. Zur zeitlichen Ordnung von Waffengräbern der jüngeren römischen Kaizerzeit in Norwegen // Beiträge zu römischer und barbarischer Bewaffung in den ersten vier nachchristichen Jahrunderten. Lublin; Marburg, Bemmann J., Hahne G. Waffenführende Grabinventare der jüngeren römischen Kaiserzeit und Völkerwanderungszeit in Skandinavien. Studie zur zeitlichen Ordnung anhand der norwegischen Funde / Bericht der Römisch- Germanischen Kommission. 75. Mainz am Rhein, Dąbrowska T. Późne odmiany fibul silnie profilowanych w Polsce // Wiadomości Archeologiczne LIII. Doppelfeld O. Rőmisches und Fränkisches Glas in Köln. Köln, Droberjar E. On Certain Amulet Pendants in the Shape of Miniature Pots, Occurring Between the Black Sea Region (Pontus Euxinus) and Scandinavia // Inter Ambo Maria. Contacts between Scandinavia and Crimea in the Roman Period. Kristiansand; Simferopol, Eggers H. Das romische Import in freien Germanien. Hamburg, Gavritukhin I. On the Study of Double-Plate Fibulas of the First Subgroup // Jósa András Múzeum Évkönyve. XIV. Nyiregyháza, Godtowski K. Zmiany w uzbrojeniu ludności kultury przeworskiej w okresie wptywów rzymskich // Arma et ollae. Studia dedykowane Profesorowi Andrzejowi Nadolskiemu w 70 rocznicę urodzin i 45 rocznicę pracy naukowej. Łódź, Ilkjaer J. Illerup Adal. Bd. 1. Die Lanzen und Speere. Textband // Jutland archaeological societt publications XXV: 1. Jahn M. Die Bewaffnung der Germanen. Wűrzburg, Kazanski M. Les éperons, les umbo, les manipules de boucliers et les haches de l epoque romaine tardive dans la region pontique: origine et diffusion // Beiträge zu römischer und barbarischer Bewaffung in den ersten vier nachchristichen Jahrunderten. Lublin; Marburg, Kazanski M. Les antiquites germaniques de l`epoque romaine tardive en Crimee et dans la region de la Mer d`azov // Ancient West and East /2. Khrapunov I. On the contacts between the populations of the Crimea and the Carpathian Basin in the Late Roman Period // International Connections of the Barbarians of the Carpathian Basin in the 1 st 5th centuries A. D. Aszód; Nyíregyháza, Khrapunov I. N. A New Account on the Contacts between the Sarmatians and the Germans in Crimea (on the materials of the excavations on the cemetery of Neyzats in 2001) // Europa Barbarica. / Monumenta studia Go thica. IV. Lublin, Kieferling G. Bemerkungen zu Äxten der römischen Kaiserzeit und der frűhen Vőlkerwanderungszeit im mitteleuropäischen Barbaricum // Beiträge zu römischer und barbarischer Bewaffung in den ersten vier nachchristichen Jahrunderten. Lublin; Marburg, Kokowski A. L art militaire des Goths à l époque romaine tardive (d apres les données archéologiques // L armee romaine et les barbares du III au VII siecles. [no place], Kokowski A. O tak zwanych blaszanych fibulach z pótokrągtą plyta na gtówce i rombowatą nóżką // Studia Gothica. I. Lublin, Kokowski A. Metalowe wisiorki w ksztatcie topora na terenie Barbaricum na pótnoc i pótnocny-wschód od limesu rzymskiego, w okresie rzymskim i we wczesnym okresie wędrówek ludów // 20 lat arrcheologii w Masłomęczu. I. Weterani. Lublin, 1998.

192 192 Igor KHRAPUNOV Kokowski A. Przyczynek do historii kontaktów germanów i sarmatów w świetle badań archeologicznych // Sarmaci i germanie. Lublin, Kontny B. Uzbrojenie kultury przeworskiej w okresie wpływów rzymskich i początkach okresu wędrówek ludów // Wandalowie. Strażnicy bursztynowego szlaku. Lublin; Warszawa, Pánczál S., Dobos A. Facet Cut Glass Vessels of the Late 3 th to 5 th Century AD. Analyses of Finds from North Danubian Romania // Funerary Offerings and Votive Depositions in Europe s 1 st Millenium AD. Cultural Artefacts and Local Identities. Cluj-Napoca, Puzdrovskij A. E., Zaicev Ju. P. Frunkbestattungen des 1. Jhs n. Chr. in der Necropole Ust -Al ma, Krim // Eurasia Antiqua Rau G. Korpergraber mit Glasbeigaben des 4. nachchristlichen Jahrhunderts im Oder Weichel Raum //Acta Praehistorica et Archaeologica Ščukin M. B. A propos des contacts militaires entre les Sarmates et les Germains a l époque romaine (d aprés et spécialement les umbo de boucliers et les lances) // L armée romaine et les barbares du IIIe au VIIe siécles. [no place], Straume E. Gläser mit Faccetenshliff aus scandinavischen Gräben des 4. und 5. Jahrhunderts n. Chr. // Universitetsforlaget. Serie B. Skrifler LXXIII. Oslo, Tempelmann-Mączyńska M. Die Perlen der rőmischen Kaiserzeit und der frűhen Phase der Vőlkerwanderungszeit im mitteleuropäischen Barbaricum. Mainz am Rhein, Thomas S. Studien zu den germanischen Kämmen der rőmischen Kaiserzeit // Arbeits- und Forschungsberichte zur sächsischen Bodendenkmalpflege. Bd. 8. Leipzig, Zieling N. Studien zu germanischen Schilden der Spătlatene- und rőmischen Kaiserzeit im freien Germanien / BAR IS Oxford, Алексеева Е. М. Античные бусы Северного Причерноморья / САИ. Г1 12. М., Амброз А. К. Фибулы юга Европейской части СССР / САИ. Д М., Арсеньева Т. М., Шелов Д. Б. Раскопки юго-западного участка Танаиса ( гг.) // Археологические памятники Нижнего Подонья. I. М., Ахмедов И. Р., Гущина И. И., Журавлёв Д. В. Богатое погребение II в. н. э. из могильника Бельбек IV // Поздние скифы Крыма. М., Бабенчиков В. П. Чорнорiченський могильник // Археологiчнi пам ятки УРСР XIII. Бажан И. А., Каргапольцев С. Ю. Об одной категории украшений-амулетов римского времени в Восточной Европе // СА Безуглов С. И. Позднесарматские мечи (по материалам Подонья) // Сарматы и их соседи на Дону. Ростовна-Дону, Беспалый Е. И. Погребения позднесарматского времени у г. Азова // СА Богданова Н. А., Гущина И. И., Лобода И. И. Могильник Скалистое III в юго-западном Крыму (I III вв.) // СА Веймарн Е. В. Археологiчнi роботи в районе Iнкермана // Археологiчнi пам ятки УРСР XIII. Велков И. Нови могилни находки // ИБАИ V. Гавритухин И. О. Финал традиций культур римского времени в восточном Прикарпатье // Die spätrömishe Kaiserzeit und die frühe Völkerwanderungzeit in Mittel- und Osteuropa. Lódź, Гопкало О. В. Бусы и подвески черняховской культуры. Киев, Гудкова А. В. Орнаментация черняховской сероглиняной керамики с памятников Буджака // Памятники древнего искусства северо-западного Причерноморья. Киев, Гудкова А. В., Фокеев М. М. Земледельцы и кочевники в низовьях Дуная I IV вв. н. э. Киев, Гущина И. И., Засецкая И. П. Золотое кладбище римской эпохи в Прикубанье. СПб., Зайцев Ю. П., Волошинов А. А., Кюнельт Э., Масякин В. В., Мордвинцева В. И., Фирсов К. Б., Флесс Ф. Позднескифский некрополь Заветное (Алма-Кермен) 1 3 вв. н. э. в юго-западном Крыму. Раскопки 2004 г. // Древняя Таврика. Симферополь, Зайцев Ю. П., Мордвинцева В. И. Исследование могильника у с. Суворово в 2001 г. // МАИЭТ X. Зубарь В. М., Мещеряков В. Ф. Некоторые данные о верованиях населения Херсонеса (по материалам некрополя первых веков н. э.) // Население и культура Крыма в первые века н. э. Киев, 1983.

193 Germanic Artefacts in the Cemetery of Neyzats 193 Казанский М. Готы на Боспоре Киммерийском // Сто лет черняховской культуре. Киев, Казанский М. Германцы в Юго-Западном Крыму в позднеримское время и в эпоху Великого переселения народов // Готы и Рим. Киев, Каргапольцев С. Ю., Бажан И. А. Умбоны щитов и боевые топоры римского времени (к вопросу о хронологии и исторической интерпретации) // ПАВ Кропоткин В. В. Римские импортные изделия в Восточной Европе (II в. до н. э. V в. н. э.) / САИ. В1 17. М., Кропоткин В. В. Черняховская культура и Северное Причерноморье // Проблемы советской археологии. М., Магомедов Б. Черняховская культура. Проблема этноса. Lublin, Магомедов Б. В., Левада М. Е. Оружие черняховской культуры // МАИЭТ V. Мастыкова А. В. О распространении янтарных грибовидных бус-подвесок позднеримского времени на юге Восточной Европы и в Закавказье // 100 лет черняховской культуре. Киев, Мастыкова А. В. Женский костюм Центрального и Западного Предкавказья в конце IV середине VI в. н. э. М., Мыц В. Л., Лысенко А. В., Щукин М. Б., Шаров О. В. Чатыр-Даг некрополь римской эпохи в Крыму. СПб., Неневоля И. И., Волошинов А. А. Два комплекса IV в. н. э. на могильнике Краснозорье // Поздние скифы Крыма. М., Петраускас О. В., Пастернак В. В. Скляні посудини могильника черняхівської культури Велика Бугаївка в Середньому Подніпров ї // Археологія Пиоро И. С. Крымская Готия. Киев, Пуздровский А. Е. Крымская Скифия II в. до н. э. III в. н. э. Погребальные памятники. Симферополь, Пуздровский А. Е., Зайцев Ю. П., Неневоля И. И. Новые памятники III IV вв. н. э. в юго-западном Крыму // МАИЭТ VIII. Россохацкий А. А. Стеклянные кубки из памятников черняховской культуры степной зоны междуречья Днестра и Дуная // Новые исследования по археологии Северного Причерноморья. Киев, Симоненко А. В. Сарматские всадники Северного Причерноморья СПб., Стоянова А. А. Бусы и подвески из могильника Нейзац (по материалам раскопок гг.) // БИ V. Cтоянова А. А. Металеві підвіски у формі сокирок із Криму // Археологія. 2005a. 2. Стоянова А. А. Подбойная могила 193 из могильника Нейзац в Центральном Крыму // МАИЭТ. 2005b. XI. Сымонович Э. А. Орнаментация черняховской керамики // МИА Симонович Э. А. О керамике черняховского типа в Крыму // Археологiя Тиліщак В. С. Знахідки кольчужного плетива на Чернелево-Руському могильнику // Археологія і давня історія України. Вип. 2. Київ, Тиліщак В. С. Про повторне використання фрагментів кольчужного плетива у центральної та східної Евро пі // Археологія Хазанов А. М. Очерки военного дела сарматов. М., Храпунов И. Н. О позднесарматской археологической культуре в Крыму // Проблемы скифо-сарматской археологии Северного Причерноморья (К 100-летию Б. Н. Гракова). Запорожье, Храпунов И. Н. Могильник Дружное (III IV вв. нашей эры). Lublin, Храпунов И. Н. Новые данные о сармато-германских контактах в Крыму (по материалам раскопок могильника Нейзац) // БИ III. Храпунов И. Н. Последние погребения в могильнике Нейзац // МАИЭТ XI. Храпунов И. Н. Погребения детей в могильнике Нейзац // МАИЭТ XII/1. Храпунов И. Н. Две могилы с погребениями женщин из некрополя Нейзац // МАИЭТ XIII.

194 194 Igor KHRAPUNOV Храпунов И. Н. Склеп IV в. н. э. из могильника Нейзац // ПИФК. XXI. М.; Магнитогорск; Новоси бирск, Храпунов И. Н. Оружие из могильника Нейзац // Terra Barbarica. Lodz; Warsawa, 2010a. Храпунов И. Н. Изделия сплетённые из железных колец // БИ. 2010b. XXIII. Храпунов И. Н. Некоторые итоги исследований могильника Нейзац // Исследования могильника Нейзац. Симферополь, 2011a. Храпунов И. Н. Склеп с погребениями III IV вв. н. э. из могильника Нейзац. Симферополь, 2011b. Шабанов С. Б. Стеклянные сосуды из могильника Нейзац (по материалам раскопок гг.) // МАИЭТ XVII. Юрочкин В. Ю. Черняховская керамика юго-западного Крыма // ХСб. X. Севастополь, Юрочкин В. Ю., Труфанов А. А. Позднеантичный погребальный комплекс в низовьях реки Качи // ХСб. XII. Севастополь, Игорь ХРАПУНОВ Германские вещи из могильника Нейзац Резюме Различные категории изделий германского происхождения, найденные в Крыму, не раз привлекали внимание исследователей. Их изучали, как правило, в контексте комплексной проблемы времени и обстоятельств проникновения германских племен, в первую очередь готов, в Крым, а также расселения их на территории полуострова. Могильник Нейзац расположен в центре крымских предгорий, приблизительно в 20 км восточнее Симферополя. За время раскопок, ведущихся с 1996 г., исследовано 518 погребальных сооружений, включая склепы, подбойные и грунтовые могилы. Датируется могильник II IV вв. н. э. При раскопках могильника Нейзац найдено некоторое количество вещей тех типов, что, как полагают многие исследователи, попали в Крым из ареала германских культур. В качестве источников предметов германского облика, как правило, рассматривается черняховская культура, ближайшая к Крыму из оставленных германскими племенами. Анализ находок показал, что подвески в виде топориков, подвески в виде ведерок, янтарные восьмерковидные подвески или какая-то их часть могли попасть в Крым вместе с германскими племенами, но могли быть изготовлены и на месте. Скорее всего, из ареала черняховской культуры привезли в крымские предгорья стеклянную посуду и двупластинчатую фибулу. Почти наверняка носителями черняховской культуры были изготовлены оказавшиеся в Крыму костяные гребни и браслет из рога. «Воинские» фибулы, наверное, изготавливались на полуострове, но, возможно, по черняховским образцам. Вероятно, черняховское происхождение имеет сероглиняная гончарная посуда, но то обстоятельство, что один из самых распространенных в Крыму ее типов не имеет аналогий в черняховской культуре, позволяет предположить наличие еще какого-то гончарного центра, поставлявшего свою продукцию в Крым. Топор и щит с металлическими деталями могли быть военными трофеями. Причем для щита более вероятно пшеворское, а не черняховское происхождение. На связи с пшеворской культурой, причем очень ранние, наладившиеся еще до середины III в. н. э., указывают и обрывки изделий из железных колец, а также, возможно, обломок фибулы типа А Главная проблема, которая возникала перед всеми, кто занимался изучением германских древностей в Крыму, заключалась в том, как вещи из ареала германских культур попадали на полуостров. Исследования могильника Нейзац не дали ответа на этот вопрос. Нет ни одного погребения, совершенного по обряду какого-либо из германских племен, нет ни одной могилы, где были бы сконцентрированы германские вещи. Они рассеяны в качестве единичных включений в огромное количест-

195 Germanic Artefacts in the Cemetery of Neyzats во сарматского погребального инвентаря. Следовательно, нет никаких оснований полагать, что среди сарматов, оставивших могильник Нейзац, хоронили германцы. О том, как попали германские вещи в Крым, в результате военных, торговых или культурных контактов, 195 где происходили эти контакты, остается только гадать. Любое из этих предположений теоретически правомерно, но ни одно из них нельзя подкрепить анализом конкретных источников, полученных при раскопках могильника Нейзац.

196 196 Bartosz KONTNY New Traces to Solve the Riddle: weapons from Chatyr-Dag in the light of current research 1 The cemetery of Chatyr-Dag had been drawing scholars attention from the first mentions concerning graves and their furnishing. One of the most disputable problems is the appearance of cremation graves in stone cists equipped with weapons, frequently ritually bent. Is was generally accepted that they express not local society, but rather newcomers from the north, most probably Goths who, according to ancient written sources, invaded the Crimea in the mid-third century AD (Пиоро 1990; Айбабин 1996: ; 1999: ; Храпунов 1999: 153; Aibabin 2003). However direct connection with the Goths cultural circle seems improbable as weapons in graves are absent here with several exceptions from the Chernyakhov culture (Магомедов, Левада 1996) and even less from the Wielbark culture (Kaczanowski, Zaborowski 1988; Kontny 2006; Kontny, Natuniewicz-Sekuła 2007; Kontny 2008). Therefore another explanation was needed. 2 Peculiarity of that cemetery, i. e. cremation in stone cists, almost unknown from the Crimean Peninsula (with the exception of Ay- Todor: Блаватский 1951), 3 let particular archaeologist to link them with Roman period Norway (Kazanski 1991: 496, and after him, e. g., Айбабин 1999: 245; Mыц et al. 2006: 186). Also a few types of weapons, having analogies in Scandinavia, justified such presumption. However lately Frans- Arne Stylegar has touched that matter from the Scandinavian point of view (Stylegar 2011: ). In his opinion, bent weapons and stone cists are well known from Norway as well as the third untypical element from Chatyr-Dag cemetery, i. e. agricultural tools. Nevertheless, these traits are hardly ever found combined, like in Chatyr- Dag. Moreover, Norwegian stone cist graves with weapons are not flat as was in the case of Chatyr- Dag, but were placed in mounds. Additionally, agricultural utensils were very rare in Norwegian graves of the Roman Period, with the exception of curved knives aimed for animal hides preparation. Stylegar in conclusion doubts the Norwegian connection and points another possible area to scrutinize, i. e. Saxony, Mecklenburg and Pomerania, where such elements like agricultural implements were found in weapon graves. I personally agree with the above idea although I would like to show more possibilities and traces which has been lost in the discussion so far. 4 Grave 2 is dated from the second half of the third to the first half of the fourth century AD, although even wider borders are possible (Mыц et al. 2006: ). Apart from the amphora, serving as an urn, a sword was found here (fig. 1. 1). It was identified as a Przeworsk culture type Biborski X (Вознесенская, Левада 1999: 253; Mыц et al. 2006: ), namely because of the fact that it was bent in a way, spotted in that culture. Classification after Marcin Biborski based on a small collection, as refers to type X which is to be 1 I would like to thank many colleagues for discussions, help and remarks concerning the particular aspects of the problem I have dealt with, specifically Anna Bitner-Wróblewska, Katarzyna Czarnecka, Igor Gavritukhin, Maxim Levada, and Aleksandr Vasil yev. 2 Because of this discrepancy of data some scholars link Chatyr-Dag cemetery with generally Germanic, not particularly Goth s population, invading Crimea in the mid-third century (Khrapunov 2012: 219). 3 Therefore Chatyr-Dag is included into Ay-Todor type of cemeteries (Kazanski 1991: ); apart from the above two cemeteries there are assumptions concerning other ones however so far they are weak (Казанский 2006: 29). 4 For adequate summary of discussion on the matter of ethnicity of people utilizing the cemetery of Chatyr-Dag, see: Казанский 2006:

197 New Traces to Solve the Riddle: Weapons from Chatyr-Dag in the Light of Current Research considered here (Biborski 1978: 91 92), so it isn t apt enough to be used here any more. More modern proposition is a typology after Biborski and Jørgen Ilkjær (2006) which allows us to attribute the sword to type Nydam-Kragehul, subtype 2 (fig ). Its shape and size meet the definition of the type, even lenticular cross-section of the blade although rare was spotted on particular examples of the type (Biborski, Ilkjær 2006: 236, ). Subtype 2 was equally popular in Scandinavia and in the Przeworsk culture (single items are known also from the Rhine limes area; see fig. 1. 6) and was generally dated to phase C2, with possible prolongation of that period (Biborski, Ilkjær 2006: Tab. 32). However single element is able to blur this simple image. In the upper part of the thong there is a small hole, seemingly aimed to fasten organic parts of the handle. Such detail has been observed on subtype 2 extremely rarely 5 (Biborski, Ilkjær 2006: ) nevertheless it is not excluded. On the Sarmatian swords such detail was observed much more frequently, namely at the end of a handle (see: Хазанов 1971: табл. IX XIII) and so it was in case of swords from the Tsebelda (Tsibilium) culture in Abkhazia (Воронов, Шенкао 1982: рис , 8, 12 13, 15). Confirmation may be shown also in the Chernyakhov culture, e. g., Kompaniytsy, Kobelyaki rayon (district), grave 86, and Mogritsa, Sumy rayon (Магомедов, Левада 1996, рис. 2. 1, 5). Therefore I assume that we deal here with barbarian sword, probably of Germanic origin 6 but with possible innovations or traces of repairing being a part of Pontic zone influence. Similar swords are known also from the Crimea (Mыц et al. 2006: 117). In the interment also two lance-heads were found. First of them (fig. 2. 1) belongs to Type III after Vytautas Kazakevichyus (Kазакявичюс 1988), so-called sword-like ones, typical of the territory of eastern Lithuania (fig ) and dated to the late fifth seventh centuries AD (Kазакявичюс 1988: 41 42, map VII). Its chronology bases on unstable 197 grounds (Kurila 2007: 299), so in my opinion they might have appeared earlier, even in the Late Roman period. It seems probable if we take into consideration very similar forms typical of the latest stage of the Luboszyce culture, i. e. Dresden-Dobritz, Stadtkreis (region) Dresden, grave 1 (Meyer 1971: 50, fig ) from the Early Migration period (fig ). 7 Naturally, its reconstruction does not seem utterly trustful but it resembles sword-like lance-heads very well. Lately also the item from the Przeworsk culture appeared: Olsztyn, powiat Olsztyn, loose find. 8 Moreover the idea of copying the image of a sword in lance-head s shape, probably having symbolic connotations as it was considered by Katarzyna Czarnecka (2010), started earlier. It is proved by single lance-heads made of broken swords, blades of which were riveted to the notched sockets. Unfortunately, such lance-heads were quite rare and scattered in different part of Barbarian world in the Late Roman and Early Migration periods. Firstly they were discovered at the Przeworsk culture cemetery at Radawa, powiat Wiązownica (fig. 2. 8) and published by Andrzej Kokowski (2000: 307, fig. 5; 2003, fig. 54), then Czarnecka found further analogies among the Przeworsk culture lance-heads (fig ) from Cierniówka, powiat Grójec (Czarnecka 2010: , fig. 1) and Tarnówko, powiat Kruszwica (Czarnecka 2010: 114, 116, fig. 3), as well as Sarmatian grave from Csongrád-Berzsenyi utca (street), Csongrád county, grave 4 (Párducz 1963: pl. II. 10), late Sarmatian or Hunnic grave from Pokrovsk, Voskhod, Engel s district, Rus sia (Синицын 1936: рис. 3; Czarnecka 2010: , fig. 6), or even from Kazakhstan (fig ) Lebedyovka, Chingirlau rayon, mound 1 (Багриков, Сенигова 1968: 82, рис ; 11; Czarnecka 2010: 120, fig. 5). I may add to the list a find from the Tsebelda culture cemetery at Tsibilium 8, grave 448/4 (fig. 2. 7), in Abkhazia (Voronov 2007: 99, fig ), made in the same technique, but most probably not of a sword s blade. Of course such state of knowledge doesn t let to 5 E. g., one sword from bog site Ejsbøl, Haderslev kommune (county) (Ørsnes 1988: pl ; 73. 7) but also the one from Budy Łańcuckie, powiat (county) Białobrzegi in the Przeworsk culture (Osiński 1923: 12, fig. 1; Biborski 1978: 89 90, 147; Koperski 2003: 126, fig. 4). There were also spotted on the blades of destructed swords from bog sites at Ejsbøl and Nydam, Sønderborg kommune (Biborski, Ilkjær 2006: 246). 6 Metallographic analysis has proved simple method of production (Вознесенская, Левада 1999: ), so it rather cannot be Roman. 7 Single find of a similar but not identical form (blade edges are not parallel) is known from sacrificial bog site at Balsmyr on Bornholm; however it seems that it is not local but Balt (Lithuanian?) element (Nørgård Jørgensen 2008: , 227; fig ). 8 Personal commitment of PhD. Katarzyna Czarnecka whom I would like to express my gratefulness.

198

199 New Traces to Solve the Riddle: Weapons from Chatyr-Dag in the Light of Current Research make ultimate statements (majority of the finds is imprecisely dated and all the Przeworsk culture ones are only loose finds). Nevertheless, it seems obvious that we deal with the eastern idea of reusing swords blades that came via Pontic zone to the Przeworsk culture most probably in the Early Migration period (see Czarnecka 2010: ). Further it came probably to the Balts areas and the Luboszyce culture but exclusively as a representation of a sword not the sword s blade as an element of the shafted weapon s head. The exclusive item from Tsibilium is dated as early as phase I/1 of the culture (Voronov 2007: 99), i. e. AD 230/ (Kazanski, Mastykova 2007: 21) so in the East the idea could have existed quite early (of course if the chronology is set adequately, as the grounds for dating seem not to be utterly trustful: it was found together with horse harness elements exclusively). The lance-head from Chatyr-Dag should be treated as a link of the presented chain of contacts from the East to the West, also quite early (if we take into consideration chronology of the sword, it should be most probably linked with phase C2, i. e. the second half of the third century AD or early fourth century A.D.). The second lance-head (fig. 4. 6) was ascribed (Mыц et al. 2006: 121) to type XX after Piotr Kaczanowski (1995) what seems to be a misunderstanding as the ratio: socket s length / total length doesn t fit the limits established for type XX; if we try to compare it with the Przeworsk culture pattern it is better to identify it as type X 9 ; more massive examples of it (close to the one in question) appeared as late as sub-phase C1b; it is similar also to type XXII from phase C1, but mainly phases C2 and C3 D (see Kaczanowski 1995: 20 21, 26) so it 199 seems most adequate to describe it as type X/XXII. Similarities to the Chernyakhov culture specimens, although possible, are hard to be proved as their number is small and they has not been published aptly enough to make any definite statements. Apart from weapons the grave furnishing contained also a sickle (fig. 3. 3). 10 This agricultural implement although well known in Barbaricum (Rodzińska-Nowak 2012: 92 97) appeared in Barbarian graves only exclusively, being characteristic for settlements (Raddatz 1993: 177; Вознесенская, Левада 1999: 262) although it was proved even for Scandinavian bog site in Vimose, former Odense county, on Funen (Engelhardt 1869: 26 27, fig ). They are almost unknown in the Crimea so the northern trace was pointed out, i. e. findings from the Tarand graves culture, e. g. Jäbara, Ida-Viru maakond (county) (Mыц et al. 2006: ), but mentioned items represent actually different form of the sickle knife / bushwhacker (see Шмидехельм 1955: pис ; 16. 3, 8). Nevertheless findings of both sickles and short scythes are proved for Iron Age Finland (Vilkuna 1934: , fig. 4), coming quite frequently (fig ) from the Early Roman period graves but also from hoards (Salo 1968: ; Kivikoski 1964: fig. 123; Kivikoski 1973: 26, pl ). In my opinion, one should underline also another area: the Dollkeim-Kovrovo culture, where both agricultural and handicrafts tools appeared in grave equipment quite frequently, both in cremations and inhumations, mostly together with weapons (fig. 3. 6). It refers also to short scythes (Raddatz 1993: fig. 30). However, there are findings of short scythes also from the other Balt cultures (fig. 3:7), e. g., from the third century on they 9 I agree with attribution proposed by Galina Voznesenskaya and Maxim Levada (Вознесенская, Левада 1999: 255). 10 The difference between short sickles and half-scythes is based on the observation that the former possess crescent or elliptical blade while the latter the straight one or close to such shape (Janowski, Kurasiński 2010: 79 80). It used also to be linked with the angle between the blade and the handle: parallel in case of the sickle and perpendicular in short scythes (see Moszyński 1929: , fig , ). However, detailed identification of particular utensil could be problematic (Janowski, Kurasiński 2010: 79 80; Rodzińska-Nowak 2012: 94) the more so that the function of both groups was almost identical. Probably also their symbolic values were very close. Fig. 1. Swords of type Nydam-Kragehul, subtype 2 (after Biborski and Ilkjær): 1 Chatyr-Dag, grave 2 (Вознесенская, Левада 1999: рис. 2); 2 5 parallels from Central and Northern Europe (Biborski, Ilkjær 2006: fig. 149): 2 Kragehul, 3 Opatów, grave 289, 4 Ejsbøl, 5 Chmielów Piaskowy, grave 7; 6 map of finds of type Nydam-Kragehul (Biborski, Ilkjær 2006: fig. 151)

200 Fig. 2. Sword-like heads of the shafted weapon (1 6) and heads of the shafted weapon with blades riveted to sockets (7 11). 1 Chatyr-Dag, grave 2 (Вознесенская, Левада 1999, рис. 5. 1); 2 Chatyr-Dag, grave 3 (Вознесенская, Левада 1999, рис. 5. 2); 3 4 Dresden-Dobritz, grave 1 (Meyer 1971, fig ); 5 6 Grigiškės-Neravai, barrow 20 (Kазакявичюс 1988, рис , 4); 7 Tsibilium 8, grave 448/4 (Voronov 2007: fig ); 8 Radawa, loose find (Kokowski 2000, fig. 5); 9 Tarnówko, loose find (Czarnecka 2010, fig. 3); 10 Cierniówka, loose find (Czarnecka 2010: fig. 1); 11 Lebedyovka, barrow 1 (Czarnecka 2010: fig. 5)

201 New Traces to Solve the Riddle: Weapons from Chatyr-Dag in the Light of Current Research became popular grave equipment element in the South-West Latvian Flat Cemetery culture (Radiņš 2007: , fig ), Proto-Semigallian barrows (Banytė-Rowell, Buža 2005: 10; fig. 38, 89), Proto-Selonian barrows (Griciuvienė, Buža 2007a: fig. 27), not mentioning in details much later, Late Migration period finds (Griciuvienė, Grižas, Buža 2005: fig. 118, 131; Radiņš 2007: pl. 743; see also Estonian finds at: Tvauri 2012: , fig ). It is stated that short scythes were known in the Balt milieu from the beginning of Christian era, generally from male graves, and sickles appeared even earlier, in the Early Iron Age, specifically in interments of Semigallian, Selonian and less often in Samogitian and Aukštainiai women although they became popular varied in form and size in the middle of the first millennium AD (Griciuvienė, Buža 2007b: 190; Tvauri 2012: 101). The appearance of short scythes in weapon graves is explained by their usefulness in making or repairing the shafts of arrows or shafted weapons (Raddatz 1993: 177) but I do not find it convincing. 11 They represent not exactly the same form as the sickles from Chatyr-Dag but first of all we cannot neglect the fact that in the Balt milieu they served as an important symbolic sign of human position in burial rite, the same as in discussed Crimean cemetery. In between we have interesting find a hoard A from Wolica Brzozowa, powiat Zamość, in eastern Poland, where together with weapons and agricultural tools, several sickles / short scythes of the Crimean type were found (Bagińska 2004: ; Niezabitowska-Wiśniewska 2009: 203, fig. 25). So far it has been dated to sub-phase D2 or the second half of the fifth century, although it is doubted by Judyta Rodzińska-Nowak, who underlines that items of possibly later chronology, i. e. horseshoes, shovel fittings and maybe a fragment of the Silesian type iron bowl were found there. Such bowls appeared in hoards with iron tools in the eighth and ninth centuries (Rodzińska-Nowak : 96). Nevertheless we should not expel the finding from our sight as surely earlier elements were included in the hoard, like iron tendril brooch. Additionally one should mention finds of the sickles from Sarmatian cemetery (fig. 3. 4) at former Go rodskoy (nowadays the territory is covered by Kras nodarsk artificial lake), grave 16 (Сазонов et al. 1995: рис ) and 21 (Сазонов et al. 1995: рис. 5. 9) both dated by the authors of the publication to the second half of the second century (Сазонов et al. 1995: 127). However known from further sites of Sarmatian-Meotian character (Сазонов et al. 1995: 119), they were definitely of a different shape than the ones from Chatyr-Dag. Much closer, although not identical sickle comes from Sarmatian / Alanic (?) grave (fig. 3. 5) at Tiszalök-Rázompuszta, Szabolcs-Szatmár-Bereg megye (county) (Istvánovits, Kulcsár, Carnap- Bornheim 2006: 93, fig. 7. 2). 12 Also Roman origin of the Chatyr-Dag sickles cannot be rejected ultimately, although it seems less probable as their handle are a simple elongation of the blade s curve but Roman ones had it out-curved falx messoria (see e. g. White 1967: 72 85, fig , pl. 7; Baumann 1983: pl. XL. 5; Gaitzsch 1985: pl. VI), bent at the greater angle falx faenaria, i. e. a scythe (White 1967: , fig. 76, 79; Zeepvat 1991: fig E) or the blade was more crescent (Curle 1901: pl. LXI. 2, 3). Therefore most probably the sickles from Chatyr-Dag were locally made. Grave 3, dated from the second half of the third to the early fourth century AD (Mыц et al. 2006: 151, 153) 13 contained a sickle (fig. 3. 2) but also a lance-head (fig. 2. 2), adze-axe (fig ) and shield fittings (fig ). The lancehead was attributed to Kaczanowski s type XX. 2 (Вознесенская, Левада 1999: 156; Mыц et al. 2006: 121, 153) dated to phase C2 what I generally accept (see Kaczanowski 1995: 25), 14 though I have to underline the extraordinary feature: faceted ring, situated at the hole of the socket. Its aim 11 It could be linked with horses, i. e., short scythes might have been used for cutting the grass for horse s fodder. One should remember that the position of a horse and a rider was quite important among the Balts (see various papers in: The Horse 2009). However, short scythes appeared also in weapon graves not furnished with spurs or horse harness (see Raddatz 1993: fig. 30) so it seems not so obvious at least for the period in question (as refers to later times sickles were connected strictly with horse graves in the East Lithuanian Barrows culture, e. g. the cemetery at Žvirbliai: Iwanowska 2006: 69 71). From the other side nomadic (thus probably not harvesting) Sarmatians placed sickles in their graves what could be an argument for the above hypothesis. 12 For further examples of Sarmatian sickles see (Istvánovits, Kulcsár, Carnap-Bornheim 2006: 106). 13 Though it seems possible, taking into consideration shield elements, I cannot exclude also later dating even to the early fifth century. 14 I agree with the attribution to the type but not the subtype as such variant as XX. 2 has not been established yet (see Kaczanowski 1995: 25).

202 Fig. 3. Chosen utensils from Chatyr-Dag and their parallels: sickles (1 5, 8 9), short scythes (6 7, 10 13), sickle knife (14) and dolabra type (15 17) tools: 1 Chatyr-Dag, grave 55 (Вознесенская, Левада 1999: рис. 8. 3); 2 Chatyr-Dag, grave 3 (Вознесенская, Левада 1999: рис. 8. 2); 3 Chatyr-Dag, grave 2 (Вознесенская, Левада 1999: рис. 8. 1); 4 Gorodskoy, grave 16 (Сазонов et al. 1995: рис ); 5 Tiszalök-Rázompuszta, (Istvánovits, Kulcsár, Carnap-Bornheim 2006: fig. 7. 2); 6 Kovrovo, grave 306 (Kulakov 2009: fig ); 7 Rucavas Mazkatuži, grave 19 (Radiņš 2007: fig ); 8 Laitila (Vilkuna 1934: fig. 4. a); 9 Saltvik (Vilkuna 1934: fig. 4. b); 10 Maaria (Vilkuna 1934: fig. 4. c); 11 Kurkijoki (Vilkuna 1934: fig. 4. d); 12 Akaa (Vilkuna 1934: fig. 4. e); 13 Maaria (Vilkuna 1934: fig. 4. f); 14 Laitila (Vilkuna 1934: fig. 4. g); 15 Chatyr-Dag, grave 3 (Вознесенская, Левада 1999: рис. 6. 1); 16 Tsibilium 1, grave 36 (Воронов, Шенкао 1982: рис ); 17 Künzing (Bishop, Coulston 2006: fig )

203 Fig. 4. Weapons from Chatyr-Dag with a single analogy: 1 Chatyr-Dag, grave 12 (Вознесенская, Левада 1999: рис. 3. 1); 2 Kilen-Balka, vault 1, burial IV (drawing: the author); 3 Chatyr-Dag, loose find (Вознесенская, Левада 1999: рис. 6. 2); 4 5 Chatyr-Dag, grave 3 (Mыц et al. 2006: табл ); 6 Chatyr-Dag, grave 2 (Вознесенская, Левада 1999: рис. 4. 2); 7 Chatyr-Dag, grave 20 (Mыц et al. 2006: табл ); 8 Chatyr-Dag, grave 55 (Mыц et al. 2006: табл ); 9 Chatyr-Dag, loose find (Mыц et al. 2006: табл. 15. Б. 5)

204 204 Bartosz KONTNY was probably to strengthen the fastening but also to adorn. Such ring was proved for another heads of the shafted weapon from Chatyr-Dag, i. e. the one from grave 12 (fig. 5. 1) and non-faceted form from grave 2 (the sword-like head; fig. 2. 1), but also on mentioned above heads made in peculiar way from Radawa and Cierniówka (fig. 2. 8, 10), what links them with the Przeworsk culture milieu as well as on one from Lebedyovka, barrow 1 (fig ). Another strange element is an adze-axe implement (fig ), identical with Roman tool, used by military forces, i. e. dolabra (fig ). It was used for various purposes like ditch digging, clearing scrub, sometimes even for fighting (as happened in the revolt of Florus and Sacrovir, when legionaries used them to hack the rebellious gladiators), starting from Republican period up to the third century AD (Bishop, Coulston 2006: 69, 117, 185). It is similar in form to modern forester s axe what might indicate that upturned pick-end was also used for rolling logs, a useful action during campaigns (Fuentes 1991: 73 74). Such pickaxes (fig ) were found also in the Tsebelda culture (Воронов, Шенкао 1982: 128, рис , 18 19) and in north Pontic zone (Прохорова, Гугуев 1992: рис. 3. 4; Вознесенская, Левада 1999: 256), probably influenced by Roman prototypes. Here we probably deal with Barbarian items as they seem to be much wider than the Roman ones (see White 1967: 62, fig ; Bishop, Coulston 2006: fig ; ; ). 15 Shield boss (fig. 4. 5) should be ascribed to Norbert Zieling s type K (Zieling 1989), having traits of both sub-type K1 (straight collar) and K2 (convex upper part, narrow brim). They are confirmed in the Przeworsk culture, Luboszyce culture and Scandinavia as type Ilkjær 8ad; single item comes from Carpathian basin (Zieling 1989: ; Ilkjær 2001: fig. 298). In Scandinavia they were typical of weapon group 7 up to group 12, i. e. from C1b / C2 up to D1(see Ilkjær 1990: fig. 200; Ilkjær 2001: 311). In the Przeworsk culture they were characteristic of Kazimierz Godłowski s weapon group 7b, i. e. from C1b / C2 up to C3 D1 (Godłowski 1994: fig. 1, 5). This form was also quite well represented in the Pontic zone: in the Chernyakhov culture (Магомедов, Левада 1996: , рис. 4. 1, 3, 5, 7, 9), but also in the Crimea, e. g. Ozyornoe III, Bakhchisaray district, vault 2 (Лобода 1977: 245, рис ; Kazanski 1994: , fig ), and in the Tsebelda culture (Kazanski 1994: , fig , 25; ; ; Воронов, Шенкао 1982: рис. 5. 9). They were spotted also in the Balt milieu, mostly the Dollkeim-Kovrovo culture, e. g. Gora Velikanov (Hüneberg), Zelenogradsk rayon, grave 31k (Kулаков 1998: рис ; Радюш, Скворцов 2008: 134 wrongly as type Zieling L), Bol shoye Isakovo, Gur yevsk rayon, grave 223, 16 Pervomayskoe (ex-warnikam), Ladushkin rayon, grave 21 (Feliks Jakobson s heritage 17 ), grave 84 close rather to type Zieling L (also Jakobson s heritage, see above), Kovrovo (ex- Dollkeim), Zelenogradsk rayon, graves 259 and 362 (Kulakov 2009: fig. 205; Радюш, Скворцов 2008: 133), ex-grebieten, Zelenogradsk rayon, grave 43 (Heydeck 1888), Yaroslavskoye (ex-schlak- 15 Roman dolabrae from the turn of the ages are known from Crimean sacrificial find at Gursufskoye Sedlo mountain pass (personal commitment: Mariya Novichenkova, Institute of Archaeology of the Ukrainian National Academy of Sciences, whom I would like to thank for the information). 16 Personal commitment Konstantin Skvortsov, whom I would like to thank. 17 Files of Jakobson are collected in the National History Museum of Latvia (see Archeologiczne 2011). Fig. 5. Conical head of the shafted weapon from Chatyr-Dag and its parallels: 1 Chatyr-Dag, grave 12 (Вознесенская, Левада 1999: рис. 5. 1); 2 3 Shapka, Monetnyy Kholm, loose finds (Воронов, Шенкао 1982: рис ); 4 Cobusca Veche (Магомедов, Левада 1996: рис ); 5 Budeşti (Магомедов, Левада 1996: рис ); 6 Łabapa, grave 63 (drawing: the author); 7 Kamieńczyk, grave 297 (Dąbrowska 1997: pl. CXXXVII. 2); 8 Nadkole, grave 49 (Andrzejowski 1998: pl. XXXVII. 8); 9 Ardanovo, loose find (Kobal 1997: fig. 3. d); Illerup, inv. no. VA, YQ, and ZF (Ilkjær 1990: fig. 39)

205 New Traces to Solve the Riddle: Weapons from Chatyr-Dag in the Light of Current Research 205

206 206 Bartosz KONTNY alken), Zelenogradsk rayon, loose find (Martin Jahn s heritage 18 ). Such forms are also proved for Lithuania type Kiulkys IIB: Žviliai, Šilalė district, grave 48 (Kiulkys 2010: 46, 48, ; fig. 9) and Peršaukštis-Kasčiukai, Švenčionys district, barrow 4, grave 3 (Kiulkys 2010: 46, 48, 107; fig. 8). It seems possible that they had their impact on the popularity of specifically Lithuanian type Kiulkys VD (see Kiulkys 2010: 58 62, fig. 32). So this type was common in the wide zone from Scandinavia via Balt area, territory of Poland, through the Chernyakhov culture to the Crimea and Abkhazia. Umbos were not identical in this wide zone but they expressed the same ideal. The shield grip (fig. 4. 4) belongs to type Zieling X, known from the Chernyakhov culture (Магомедов, Левада 1996: 307, рис ), the Crimea, e. g. Ay-Todor, Yalta area, grave 11, Ozyornoye, Bakhchisaray area, vault 2 (Лобода 1977: 245, рис ), Kerch, Lenino rayon, grave 181 / 1902 (Kazanski 1991: fig ; Kazanski 1994: 479, fig ), and close forms also from the Tsebelda culture (Kazanski, Mastykova 2007: fig ; ). It was one of the most widely spread forms, typical of the whole Germanic world and confirmed even in Finland end Estonia (Kazanski 1994: , fig ); single item comes from Romania (Zieling 1989: 226). Naturally they were very popular in the Przeworsk culture in Godłowski s weapon group 7b (Godłowski 1994: fig , 67) and in Scandinavia, so called type Ilkjær 5c which is an equivalent of the above forms. It is proved for quite a wide chronological range from sub-phase C1b to C3, i. e. in weapon groups 6 11 (see Ilkjær 1990: fig. 200; Ilkjær 2001: ). Additionally they are documented in the Dollkeim-Kovrovo culture, e. g. Bol shoye Isakovo, grave 65 (Радюш, Скворцов 2008: рис. 5. 7), although so far they are no evidences of their existence in the Sudovian culture or territory of Lithuania (the Bogaczewo culture area cannot be taken into consideration as almost no weapons are known here starting from phase C2 see Kontny 2009). Further disputable interment from Chatyr-Dag is grave 12, with a sword of type Khazanov V (1971) or Maeotian type (fig. 4. 1). 19 Such forms were very popular in the Crimea (fig. 4. 2) and the North Caucasus, known also in the Chernyakhov culture, and steppe zone of Eastern Europe. Single finds are proved for Pannonia, the Tisza basin, and even Gallia. The westernmost ones are linked with Alanic migrations (Левада 2013: 179, рис. 2 4). They appeared in the late third century AD and lasted till the seventh century AD, although Voznesenskaya and Levada doubt the seventh century chronology (Вознесенская, Левада 1999: 254). The earliest and the latest items come from eastern Pontic zone, namely the Caucasus / Western Caspian area. Swords and daggers with notched upper part of the blade from the Crimean peninsula are dated to the fourth century AD (see Koнтны, Савеля 2006: ; Левада 2006; Левада 2013: ). Another element from the grave is a head of the shafted weapon of elongated regular pyramid with octagon base (fig. 5. 1). Probably it was an element of a javelin. Although such proposition occurred, I do not see any parallels to Roman javelins (Вознесенская, Левада 1999: ) 20 or projectiles of ballista (Mыц et al. 2006: ). 21 It seems that we deal with the form of conically shaped heads. Such forms appeared occasionally in the Przeworsk culture (fig ) as early as the Late Pre-Roman period, and then in the Younger and Late Roman periods (Kontny 1999: , pl. 60). The later ones became popular also among the Bogaczewo culture (fig. 5. 6): Bogaczewo-Kula (ex-bogatzewen or Kullabrücke), powiat Giżycko, grave 21, two items from Gąsior (ex-jaskowska See or Gonschor), powiat Ruciane-Nida, grave 61 (Schmiedehelm 2011: 27, pl. XIX. 8, fig. 2608) and grave 235 (Schmiedehelm 2011: 48), Radzieje (ex-rosengarten), powiat Węgorzewo, grave 56 (Kon tny 2007: 93 94, fig. 6. b c, table 1, with further literature), Łabapa (ex-labap), powiat Węgorzewo, grave 63 (La Baume 1941: pl. 30 center; Prussia Archive in the collection of the Museum of Prehistory and Early History, Berlin: PM-IXd1.1950/2.1 9; inv. no of former Prussia Museum ; the find from Prussia Sammlung is now housed in the same Museum), Kosewo I (ex-kossewen), powiat Mrągowo, gra- 18 Files of Jahn are collected in the Institute of Archaeology, University of Warsaw. 19 The interment is dated from the late third to the first half of the fourth century AD what I generally accept. 20 Similar forms are known from the cemetery on the Rhine at Wederath-Belginum where also soldiers of auxilia were buried (Haffner 1974: 45; pl ), but such forms were simpler, dated to much earlier period and could have been local forms of Treverii people (see Waurick 1994: 14 15, fig. 11A. 10; 12B.d). 21 It is too ornamental to be used in that way.

207 New Traces to Solve the Riddle: Weapons from Chatyr-Dag in the Light of Current Research ve 201 (Prussia Archive: PM-A 1848/1. 128, PM-A 1848/3.181) but also the Sudovian culture (Szurpiły, powiat Jeleniewo, loose find), and the Dollkeim- Kovrovo: Yaroslavskoe (ex-schlakalken), Zelenogradsk rayon, grave 13 (Jahn s heritage, see above), Kotel nikovo (ex-warengen), Zelenogradsk rayon, loose find (Prussia-Sammlung Berlin, inv. no PM V b), Kovrovo, grave 16 (Jahn s heritage; Nowakowski 1996: pl. 5. 5), Zhukovskoye, Gur yevsk rayon, grave (?) I (Bez zenberger 1900: , fig. 39; Gaerte 1929: fig f; Jahn s heritage). Part of them could be treated as a lanceshoe (ferrule) but at least items from Łabapa, Gąsior, and Kotel nikovo seem to be javelin heads. Examplars alike appeared also in Scandinavia, i. e. type Havor (fig ) (Ilkjær 1990), in Moravia (Hrubčice, Prostějov okres [area], grave 2, and Kostelec na Hané, Prostějov okres, grave 159 (see Droberjar, Peška 1994: 299, fig , 12; with further literature); close forms may be identified among heads of type Kaczanowski XXV (Kaczanowski 1995: 27 28). Naturally, similar items (so-called pike, Russian term пика) are proved for the Chernyakhov culture see fig (Магомедов, Левада 1996: 309, fig ), the Tsebelda culture see fig (Воронов, Шенкао 1982: 127, рис ; 14 15, 18, Воро нов 1975: рис ; 30. 3; Воронов, Юшин 1979: рис , 17; 8. 40; Kazanski, Mastykova 2007: pl ) and the Crimea (Koнтны, Савеля 2006: ; Mыц et al. 2006: 121, 156). It looks like they were most popular during phases C3 D1, i. e. the fourth century AD but it was rather an intercultural trend (covering regions from Scandinavia to Abkhazia), connected with the change in tactics, i. e. fighting with armored enemy. (see: piercing capacity of a point). Therefore they were not necessarily identical. So that s why in case of Grave 12 from Chatyr-Dag we deal with an item thoroughly made and ornamental (faceted ring on the socket, identical with one from grave 3 lance-head see fig 2. 2). So it has not exact analogy but it should be rather considered as an expression of wider tendency. Then we have to do (fig. 4. 3) with the loosely found axe (Вознесенская, Левада 1999: , рис. 6. 2; Mыц et al. 2006: 123, рис. 6. 4, табл. 15B.9). This type of weapon was quite popular in 207 Pontic zone: in the Crimea (Koнтны, Савеля 2006: ), the Chernyakhovsk culture (Магомедов, Левада 1996: , рис. 5) or Abkhazia (Воронов, Шенкао 1982: , рис. 3) 22 and among Maeotian tribes Tsemdolina cemetery (Вознесенская, Левада 1999: 257; Mыц et al. 2006: 123); asymmetric axes were also known from the Luboszyce culture (Domański 1979) and Balt tribes area, namely Lithuania (Malonaitis 2008). Although it is possible to attribute the axe from Chatyr-Dag to type 2 after Yuriy Voronov and N. K. Shenkao (Воронов, Шенкао 1982: 127) or type B after Grzegorz Domański (1979) it would not be the whole truth. Similar asymmetry with relatively short head butt could be pointed out in case of Kilen-Balka, tomb 3, grave I in the Crimea from the late fourth century AD (Koнтны, Савеля 2006: 131, , рис. 5. 1), Pitschnavari in the Tsebelda culture but also in the Przeworsk culture hoard A from Wolica Brzozowa mentioned above (Bagińska 2004: ; Niezabitowska 2009: 203, fig. 25). Its asymmetry reminds also the outline of Lithuanian axes, i. e. type I, variant i after Arvydas Malonaitis (2008); the type has very wide chronology: first to eighth centuries AD (Malonaitis 2008: 296, fig. 13, 16). Finally we have three heads of the shafted weapon: identified as type Kaczanowski XIX. 2 from grave 20 see fig (Вознесенская, Левада 1999: 255, рис. 4. 3; Mыц et al. 2006: 121, 161, рис. 7. 5, табл ), type Kaczanowski XXII. 2 from grave 55 see fig (Вознесенская, Левада 1999: 255, рис. 4. 1; Mыц et al. 2006: 121, 169, рис. 7. 4, табл ), and type Kaczanow ski X loose find see fig (Mыц et al. 2006: 121, 123, рис. 7. 7, табл ). As refers to the second and third attribution I tend to agree with them 23 but I definitely oppose the first proposal: it is actually the head of type XV after Kaczanowski (see proportions and facetted socket), dated to phases B2b C1b (Kaczanowski 1995: 23). 24 Taking into consideration the chronology of the cemetery we should probably take the latest stage of the proposed time span. The same refers to dating of the head loosely found. 25 As refers to the chronology of grave 55, established for the turn of the third and fourth century AD and the early fourth century AD 22 See: Kazanski 1994: , ; Вознесенская, Левада 1999: ; Mыц et al. 2006: With the exception that Kaczanowski has not defined any variants of that type (see Kaczanowski 1995: 26). 24 Moreover there are no variants established for type XIX (see: Kaczanowski 1995: 25) hence we have to do with misunderstanding. 25 The chronology of the type embraces phases B2 C1 (Kaczanowski 1995: 20 21).

208 208 Bartosz KONTNY (Mыц et al. 2006: 169) it stands in accordance with the dating of heads type Kaczanowski XXII, i. e. C2 C3 D (Kaczanowski 1995: 26), although naturally it is not so easy to change between absolute and relative chronology. The above connections with the Przeworsk culture patterns cannot be treated as definite. This is because thorough description and scrutiny of the Chernyakhov culture heads is needed which could change that image. Concluding: we deal with many cultural traits showing connections with Balt tribes, the Przeworsk culture and Scandinavia but reaching also to the neighboring Chernyakhov culture, local Crimean elements and further to the east to east Pontic zone. Moreover all of it was found in graves of form which is non-typical for any of the above traditions. How to explain such eclectic image? Of course we have to underline that an intercultural transition belt linking Pontic zone and Scandinavia via Goths circle but also as it seems the West Balt circle and the Przeworsk culture was postulated from years. Let me remind concept of Dancheny Brangstrup horizon (Werner 1988) but lately it was deepened owing it particularly to Inter Ambo Maria conference (namely: Kazanski 2011; Magomedov 2011). Now it seems clear that it didn t confine to luxurious objects but one has to consider also military objects, e. g., military belts from Varpelev, Vallø county, and Pruszcz Gdański, powiat Pruszcz Gdański (Kontny, Ogonowska, Pietrzak, forthcoming) and weapons. Can we than pinpoint the ethnicity of people that had used Chatyr-Dag cemetery? Naturally, the migration from the north to southeast is quite possible. But can we expect exactly the same cultural model after migration? In my opinion definitely not. So called snowball mechanism makes the initiative group of migrating warriors growing bigger while wandering their path. If they had started in Scandinavia, also Balt and Przeworsk culture traits are quite understandable. Originally it could remind literary parallel, i. e. the Fellowship of the Ring from John Ronald Reuel Tolkien s The Lord of the Rings : the elf, the dwarf, hobbits and various races of people everyone with its own, peculiar weapon. Such multi-ethnical group, far from its homelands so without any strong cultural background is susceptible to changes and creation of new eclectic model. It seems to explain quite well what we observe in Chatyr-Dag also as refers to acceptance of the Pontic zone elements. It could also account for why we didn t find traces of various migrating peoples like Borani or Heruli. Possibly their homeland traits were limited almost exclusively to the name of the core of migrants (initiators of the migration) and the rest were multiple snowflakes that adhered to them creating the cultural cocktail. Bibliography Aibabin A. Zur Datierung des germanischen Einfalls auf der Krim // Kontakt Kooperation Konflikt. Germanen und Sarmaten zwischen dem 1. und dem. 4. Jahrhundert nach Christus. Neumünster, Andrzejowski J. Nadkole 2. A Cemetery of the Przeworsk Culture in Eastern Poland / Monumenta Archaeologica Barbarica. V. Kraków, Archeologiczne dziedzictwo Prus Wschodnich w archiwum Feliksa Jakobsona // Aestiorum Hereditas. II. Warszawa, Bagińska J. Wolica Brzozowa, pow. zamojski, województwo lubelskie // Wandalowie. Strażnicy bursztynowego szlaku. Lublin; Warszawa, Banytė-Rowell R., Buža Z. Proto-Semigallians, I IV cent. // Žiemgaliai. The Semigallians. Vilnius; Riga, Baumann V. H. Ferma Romană din Dobrogea. Tulcea, Bezzenberger A. Gräberfeld bei Margen, Kr. Fischhausen // Sitzungsberichte der Altertumsgesellschaft Prussia Biborski M. Miecze z okresu wpływów rzymskich na obszarze kultury przeworskiej // Materiały Archeolo giczne XVIII. Biborski M., Ilkjær J. Illerup Ådal 11. Die Schwerter. Textband / Jutland Archaeological Society Publications. 25 / 11. Moesgård, Bishop M., Coulston J. Roman Military Equipment from the Punic Wars to the Fall of Rome. 2nd ed. Oxford, Curle J. A Roman Frontier Post and its People. The Fort of Newstead in the Parish of Melrose. Glasgow, Czarnecka K. Nietypowe groty włóczni ze schyłku starożytności z ziem polskich // Wiadomości Archeologiczne. LXI ( ). Warszawa, 2010.

209 New Traces to Solve the Riddle: Weapons from Chatyr-Dag in the Light of Current Research 209 Dąbrowska T. Kamieńczyk. Ein Gräberfeld der Przeworsk-Kultur in Ostmasowien / Monumenta Archaeologica Barbarica. III. Kraków, Domański G. Kultura luboszycka między Łabą a Odrą w II IV wieku. Wrocław, Droberjar E., Peška J. Waffengräber der römischen Kaiserzeit in Mähren und die Bewaffnung aus dem Königsgrab bei Mušov // Beiträge zu römischer und barbarischer Bewaffnung in den ersten vier nachchristlichen Jahrhunderten. Marburger Kolloquium Marburg; Lublin, Engelhardt C. Vimose Fundet. Kjöbenhavn, Fuentes N. The Mule of a Soldier // Journal of Roman Military Equipment Studies Gaerte W. Urgeschichte Ostpreussens. Königsberg, Gaitzsch W. Werkzeuge und Geräte in der römischen Kaiserzeit. Eine Übersicht // Aufstieg und Niedergang der römischen Welt. II. Berlin; New York, Godłowski K. Die Chronologie der germanischen Waffengräber in der jüngeren und späten Kaiserzeit // Beiträge zu römischer und barbarischer Bewaffnung in den ersten vier nachchristlichen Jahrhunderten. Marburger Kolloquium Marburg; Lublin, Griciuvienė E., Buža Z. Selonia and Selonians according to the burial data // Sėliai. The Selonians. Vilnius; Riga, 2007a. Griciuvienė E., Buža Z. Selonian finds // Sėliai. The Selonians. Vilnius; Riga, 2007b. Griciuvienė E., Grižas G., Buža Z. Semigallians, V XIII cent. // Žiemgaliai. The Semigallians. Vilnius; Riga, Haffner A. Das keltisch-römische Gräberfeld von Wederath-Belginum. 2. Gräber , ausgegraben 1956/1957 // Trierer Grabungen und Forschungen. VI / 2. Mainz, Heydeck J. Der südliche Theil des Gräberfeldes von Grebieten // Sitzungsberichte der Altertumsgesellschaft Prussia Ilkjær J. Illerup Ådal 1. Die Lanzen und Speere. Textband // Jutland Archaeological Society Publications. 25/1. Århus, Ilkjær J. Illerup Ådal. 9. Die Schilde. Textband // Jutland Archaeological Society Publications. 25/9. Århus, Istvánovits E., Kulcsár V., Carnap-Bornheim C. von. The barbarian warrior burial from Tiszalök-Rázompuszta in the 3 rd century history of the upper Tisza region // Archäologisches Korrespondenzblatt /1. Iwanowska G. Cmentarzysko kurhanowe w Żwirblach pod Wilnem. Wprowadzenie w problematykę. Katalog. Warszawa, Janowski A., Kurasiński T. Rolnik, wojownik czy odmieniec? Próba interpretacji obecności sierpów w grobach wczesnośredniowiecznych na terenie ziem polskich // Wymiary inności. Nietypowe zjawiska w obrzędowości pogrzebowej od pradziejów po czasy nowożytne / Acta Archaeologica Lodziensia. 56. Łódź, Kaczanowski P. Klasyfikacja grotów broni drzewcowej kultury przeworskiej z okresu rzymskiego / Klasyfikacje zabytków archeologicznych. I. Kraków, Kaczanowski P., Zaborowski J. Bemerkungen über die Bewaffnung der Bevölkerung der Wielbark-Kultur // Kultura wielbarska w młodszym okresie rzymskim. II. Lublin, Kazanski M. Contribution a l histoire de la défense de la frontiere pontique au Bas-Empire // Travaux et Mé moires Kazanski M. Les éperons, les umbo, les manipules de boucliers et les haches de l époque romaine tardive dans la region pontique: origine et diffusion // Beiträge zu römischer und barbarischer Bewaffnung in den ersten vier nachchristlichen Jahrhunderten (Marburger Kolloquium 1994). Marburg; Lublin, Kazanski M. Kishpek, Ekazhevo and Varpelev: on the problem of Pontic-Scandinavian relations in the Late Roman Period // Inter Ambo Maria. Contacts between Scandinavia and the Crimea in the Roman Period (collected papers). Kristiansand; Simferopol, Kazanski M., Mastykova A. Tsibilium. La nécropole apsile de Tsibilium (VIIe av. J.-C. VIIe ap. J.-C.). (Abkhazie, Caucase). L étude du site. II /Archaeological Studies on Late Antiquity and Early Medieval Europe [ A. D.]. III. BAR IS. 1721/II. Oxford, Khrapunov I. The Crimea in the Early Iron Age. An Ethnic History. Simferopol; Kristiansand, Kieferling G. Bemerkungen zu Äxten der römischen Kaiserzeit und der frühen Völkerwanderungszeit im mitteleuropäichen Barbaricum // Beiträge zu römischer und barbarischer Bewaffnung in den ersten vier nachchristlichen Jahrhunderten / Marburger Kolloquium Marburg; Lublin, 1994.

210 210 Bartosz KONTNY Kiulkys D. Senojo ir vidurinio geležies amžiaus skydai Lietuvoje // Archaeologia Lituana Kivikoski E. Finlands förhistoria. Helsingfors, Kivikoski E. Die Eisenzeit Finnlands. Bildwerk und Text. Helsinki, Kobal I. V. Kultura przeworska na Ukrainie Zakarpackiej // Wiadomości Archeologiczne LIII/2. Kokowski A. Polsko-niemiecki przyczynek do ratowania zabytków z Radawy w woj. podkarpackim // Materiały i Studia Rzeszowskiego Ośrodka Archeologicznego XXI. Kokowski A. Die Przeworsk-Kultur. Ein Völkerverband zwischen 200 vor Chr. und 375 nach Chr. // Die Vandalen. Die Könige. Die Eliten. Die Krieger. Die Handwerker. Nordstemmen, Kontny B. Znaleziska toków z obszaru kultury przeworskiej // Światowit n. s I (XLII) / B. Kontny B. Powracający temat. Głos w sprawie odkrycia z Żarnowca // Goci i ich sąsiedzi na Pomorzu. Koszalin, Kontny B. Najwcześniejsze elementy uzbrojenia w kulturze bogaczewskiej w świetle zewnętrznych wpływów kulturowych // Kultura bogaczewska w 20 lat później. / Materiały z konferencji Seminarium bałtyjskie. I. Warszawa, Kontny B. Breves gladii et rotunda scuta. Remarks on the Goth s weapons on the margin of Tacitus text // Germania Sarmatia. Древности Центральной и Восточной Европы эпoхи римского влияния и переселения народов. Kaлининград, Kontny B. The latest weapons in the Bogaczewo culture // The Turbulent Epoch. New Materials from the Late Roman Period and the Migration Period. II. Lublin, Kontny B., Natuniewicz-Sekuła M. A Wielbark Culture piece of weaponry? Remarks concerning the astonishing find from the cemetery at Krosno (Crossen) // Weapons, Weaponry and Man. In memoriam Vytautas Kazakevicius / Archaeologia Baltica. 8. Klaipėda, Kontny B., Ogonowska Z., Pietrzak M. Nie-rzymski trop. Niezwykły pas z Pruszcza Gdańskiego, stan. 5, grób 26 // Kontakty ponadregionalne kultury wielbarskiej. Przemiany kulturowe w okresie wpływów rzymskich na Pomorzu. Gdańsk (forthcoming). Koperski A. Zbiory archeologiczne Muzeum Narodowego Ziemi Przemyskiej w Przemyślu // Rocznik Przemyski. Archeologia XXXIX/2. Kulakov V. Dollkeim-Kovrovo, Kaliningrad Region, Russia. Research on the cemetery conducted in 1879 and / BAR IS Oxford, Kurila L. Grave of the unburied: symbolic Iron Age warrior burials in East Lithuania // Weapons, Weaponry and Man. In memoriam Vytautas Kazakevicius / Archaeologia Baltica. 8. Klaipėda, La Baume W. Vorgeschichtliche Forschung und Denkmalpflege in Ostpreuβen (1939 und 1940) // Nachrichtenblatt für Deutsche Vorzeit /3 4. Magomedov B. The Chernyakhov People s Contacts with Scandinavia and the Crimea // Inter Ambo Maria. Contacts between Scandinavia and the Crimea in the Roman Period. Kristiansand; Simferopol, Malonaitis A. Geležiniai siauraašmeniai kirviai Lietuvoje. Vilnius, Meyer E. Die germanischen Bodenfunde der spätrömischen Kaiserzeit und der frühen Völkerwanderungszeit in Sachsen // Arbeits- und Forschungsberichte zur sächsischen Bodendenkmalpflege. 6. Berlin, Moszyński K. Kultura ludowa Słowian. Część I. Kultura materialna. Kraków, Niezabitowska-Wiśniewska B. Archaeology, History and the Heruls. The Lublin Region in the Late Roman Period and the Migration Period // Światowit. Supplement Series B. Barbaricum. 8. Warszawa, Nørgård Jørgensen A. Porskjær mosefund // Jysk Arkæologisk Selskabs skrifter. 59. Højbjerg, Nowakowski W. Das Samland in der römischen Kaiserzeit und seine Verbindungen mit dem römischen Reich und der barbarischen Welt. Marburg; Warszawa, Ørsnes M. Ejsbøl I. Waffenopferfunde des Jahrh. nach Chr. // Nordiske Fortidsminder. B 11. København, Osiński K. Wykopalisko w Budach Łańcuckich z epoki młodszego okresu cesarstwa rzymskiego // Rocznik Przemyski III. Párducz M. Die ethnischen Probleme der Hunnenzeit in Ungarn / Studia archaeologica. 1. Budapest, Raddatz K. Der Wolka-See, ein Opferplatz der Römischen Kaiserzeit in Ostpreussen // Offa. Berichte und Mittei lungen zur Urgeschichte, Frühgeschichte und Mittelalterarchäologie /50.

211 New Traces to Solve the Riddle: Weapons from Chatyr-Dag in the Light of Current Research 211 Radiņš A. Epoka żelaza // Skarby starożytnej Łotwy / Treasures of Ancient Latvia. Warszawa, Rodzińska-Nowak J. Gospodarka żywnościowa ludności kultury przeworskiej // Opera Archaeologiae Iagellonicae. II. Kraków, Salo U. Die frührömische Zeit in Finnland // Suomen Muinaismuistoyhdistyken Aikakauskirja Finska Fornminnesföreningens Tidskrift. 67. Helsinki, Schmiedehelm M. Das Gräberfeld Gąsior am Jaskowska-See in Masuren. Studien zur westmasurischen Kultur der römischen Kaiserzeit. Warszawa, Stylegar F.-A. Weapon Graves in Roman and Migration Period Norway (AD 1 550) // Inter Ambo Maria. Contacts between Scandinavia and the Crimea in the Roman Period. Kristiansand; Simferopol, The Horse and Man in European Antiquity (Worldview, Burial Rites, and Military and Everyday Life) / Archaeologia Baltica. 11. Klaipeda, Tvauri A. The Migration Period, Pre-Viking Age, and Viking Age in Estonia // Estonian Archaeology. 4 Tartu, Vilkuna K. Zur Geschichte der finnischen Sicheln // Excavationes et studia. Opuscula in honorem Alfred Hackman Helsinki, Voronov Yu. Tsibilium. La nécropole apsile de Tsibilium (VIIe av. J.-C. VIIe ap. J.-C.). (Abkhazie, Caucase). Les fouilles de I // Archaeological Studies on Late Antiquity and Early Medieval Europe ( A. D.). II / BAR IS. 1721/I. Oxford, Waurick G. Zur Rüstung von frühkaiserzeitlichen Hilfstruppen und Verbündeten der Römer // Beiträge zu römischer und barbarischer Bewaffnung in den ersten vier nachchristlichen Jahrhunderten. Marburger Kolloquium Marburg; Lublin, Werner J. Dančeny und Brangstrup. Untersuchungen zur Černjachov-Kultur zwischen Sereth und Dnestr und zu den Reichtumszenter auf Fünen // Bonner Jahrbücher White K. D. Agricultural Implements of the Roman World. Cambridge, Zeepvat R. J. Roman gardens in Britain // Garden Archaeology / CBA Research Report. 78. London, Zieling N. Studien zu germanischen Schilden der Spätlatène und der römischen Kaiserzeit im freien Germanien // BAR IS Oxford, Айбабин А. И. Население Крыма в середине III IV вв. // МАИЭТ V. Айбабин А. И. О дате вторжения германцев в Крым // Сто лет черняховской культуре. Киев, Багриков Г. И., Сенигова Т. Н. Oткрытие гробниц в западном Казахстане (II IV вв.) // Известия Академии Наук Казахской CCP Блаватский В. Д. Харакс // МИА. 19/I. М., Вознесенская Г. А., Левада М. Е. Кузнечные изделия из могильника Чатыр-Даг: попытка типологического анализа и технология производства // Сто лет черняховской культуре. Киев, Воронов Ю. Н. Тайна цебельдинской долины. M., Воронов Ю. Н., Шенкао Н. К. Вооружение воинов Абхазии IV VI вв. // Древности эпохи великого переселения народов V VIII веков. М., Воронов Ю. Н., Юшин В. A. Ранний горизонт (II IV вв. н. э.) в могильниках цебельдинской культуры (Абхазия) // CА Kазакявичюс В. Оружие балтских племён II VIII веков на территори Литвы. Вильнюс, Казанский М. Германцы в Юго-Западном Крыму в позднеримское время // Готы и Рим. Киев, Koнтны Б., Савеля Д. Ю. Воины на рубежe миров: вооружение из могильника в Килен-балке // МАИЭТ XII. Кулаков В. И. Зональный орнамент на сoсудах малых форм из могильников янтарного берега // 20 lat archeologii w Masłomęczu. II. Lublin, Левада M. E. Кинжалы типа Хазанов-V в Среднем Поднепровье // Боспорские чтения. VII. Керчь, Левада M. E. О влиянии аланских военных традиций на восточногерманские народы // Крым в сарматскую эпоху (II в. до н. э. IV в. н. э.). Симферополь-Бахчисарай, Лобода И. И. Раскопки могильника Озёрное III в гг. // CА Магомедов Б. В., Левада М. Е. Оружие черняховской культуры // МАИЭТ V.

212 212 Bartosz KONTNY Мыц В. Л., Лысенко A. B., Щукин M. B., Шаров O. B. Чатыр-Даг некрополь римской эпохи в Крыму. СПб., Пиоро И. С. Крымская Готия (очерки этнической истории населения Крыма в позднеримский период и раннее средневековье). Киев, Прохорова T. A., Гугуев B. K. Богатое caрматское погребение в кургане 10 Кобякского могильника // РА Радюш O., Скворцов K. Находки деталей щитов в ареале самбийско-натангийской культуры // Germania Sarmatia. Древности Центральной и Восточной Европы эпoхи римского влияния и переселения народов. Kaлининград, Сазонов A. A., Спасовский Ю. Н., Сахтарьек 3. Н., Тов A. A. Новые материалы могильника первых веков нашей эры близ хутора Городского // Aрхеология Aдыгеи. Maйкоп, Синицын И. В. Позднесарматские погребения Нижнего Поволжья. Саратов, Сокольский И. И. Боспорские мечи // MИА. 33. М., Хазанов А. М. Очерки военного дела сарматов. М., Храпунов И. Н. О населении Крыма в позднеримское время (по материалам могильника Дружное) // РА Шмидехельм M. Х. Археологические памятники периода разложения родового строя на северо-востоке Эстонии (V в. до н. э. V в. н. э.). Таллин, Бартош КОНТНЫ Новые возможности для решения загадки. Оружие из Чатыр-Дага в свете последних исследований Резюме В статье рассмотрены предметы вооружения из могильника Чатыр-Даг, к которым не раз обращались исследователи. Автор приводит ряд новых, ранее не использованных аналогий, связанных главным образом с северо-восточной Европой, точнее с балтскими племенами. Естественно, фиксируются и пшеворские элементы, а также сарматские и характерные для черняховской культуры, и, наконец, указывающие на восточно-понтийскую зону, в том числе на Кавказ. Любопытно, что, хотя и отсутствуют черты, связанные исключительно со Скандинавией, некоторые элементы были широко распространены, в том числе и на скандинавских землях. Смешение разнокультурных элементов указывает на существование линии межкультурных связей, соединявшей Понтийскую зону со Скандинавией через круг готских памятников, но так же и, как кажется, через круг западнобалтских древностей и пшеворскую культуру. Хотя о таком взаимодействии говорилось уже давно, в частности, И. Вернером, предложившим идею горизонта «Данчены Брангструп», сегодня становится ясно, что оно не ограничивалось богатыми вещами, как считалось ранее, но включало и предметы воинского снаряжения, например, боевые пояса и оружие. Этническую принадлежность населения, оставившего некрополь, трудно определить точно. Соглашаясь с теми исследователями, которые полагают, что могильник создали переселенцы с севера, автор предлагает вспомнить о так называемом «механизме снежного кома», в соответствии с которым начальная группа переселенцев-воинов увеличивается по мере продвижения вперёд. Если это движение началось в Скандинавии, нетрудно понять, откуда взялись черты балтской и пшеворской культуры. Изначально можно вспомнить литературную параллель братство кольца из романа «Властелин колец» Дж. Р. Р. Толкиена, объединившее воинов разных рас, каждый из которых имел собственное вооружение. Такая полиэтничная группа, оказавшаяся вдали от родины и не имевшая крепкого культурного фона, легко менялась, формируя новую эклектичную модель. Кажется, подобная ситуация хорошо объясняет то, что мы наблюдаем на Чатыр-Даге, в том числе и заимствование причерноморских элементов.

213 213 Maxim LEVADA Sösdala: The Problem of Singling out an Artistic Style 1 The cause to write this paper is the discussion occurring at the conference Inter Ambo Maria: Northern Barbarians from Scandinavia towards the Black Sea in Gaspra, Crimea, in October 2012 after the lecture presented by Charlotte Fabech and Ulf Näsman discussing the famous find in Sösdala (Fabech, Näsman 2012). The lecturers did not agree with my idea that the artefacts in the Sösdala style were produced by northern barbaric artisans (Fabech, Näsman 2012: 75). In their opinion, ornaments of this circle are made by Romans in the Imperial territory. What I would like to say right now, my conclusion in question concerns only one group of the finds in the mentioned style, and namely silver ornaments decorated with three jewellery techniques at once: 1) fine stamp or / and cutting tool, 2) nielloing, and 3) mercury-gilding (Levada 2011). However, now I think that all the other finds in the Sösdala style should be considered works of barbaric goldsmiths, leaving the problem of local production centres behind our discussion. Anyway, the area where this style initially appeared should be localised in Scandinavia. The core of the discussion is excellently defined by a point, which was put by Anna Bitner- Wróblewska more than ten years ago. I think it necessary to remind it: On the one hand there is a horizon of specific artefacts, on the other, a largely undefined ornamental style (Bitner-Wróblewska 2001: 89). Therefore, the problem, in my point of view, is that we should isolate ourselves as far as possible from the features of the chronological horizon Kachin Sösdala Untersiebenbrunn and concentrate our attention on artistic specificities of the Sösdala style, disregarding difficulties with singling it out. My analysis starts from the supposition that artisan of high level, since the artefacts in question display extraordinary jewellery technique, works in one manner, typical only to him. 2 This conservative feature of craft production formed some artistic styles and traditions. Artistic manner changed either throughout a relatively long period of time, or under the influence of global impulses from the outside. In the latter case, artisan had to have initially a sample to follow, i. e. prototype. For the artistic style Sösdala we do not know such stylistic samples, consequently, any speculation concerning the origins, borrowing, or outside centres of craft is nonplussed. I would like to underline again that it is not the case of borrowing of jewellery techniques, but only of an artistic style. Therefore we have enough reasons to compare synchronous finds of ornaments combining the mentioned techniques, particularly fine stamping, nielloing, and mercury-gilding, in order to understand technological differences of the finds or, on the contrary, their stylistic proximity. However, to begin with, let us remember earlier Roman artefacts combining such jewellery techniques. The combination of nielloing on silver and partial gilding is well-known among luxury officers belts, forming a good example of actually Roman style (Zsolt 2010; 2011). Particularly, the ornamentation of belt set from Budapest-Pasarét 1 This paper would not have sense without a great number of illustrations, which were very difficult to collect together. Working on it, I used to recollect Karol Pieta s brilliant aphorism: Archaeology is like pornography, which is not interesting without pictures. Therefore I would like to express my deep gratitude to all those who helped me: Natal ya Turova, Oleg Radyush, Michel Kazanski, Cristian Pilet, Valéria Kulcsár, Margit Nagy, Nikita Khrapunov, Frans-Arne Stylegar, Aleksandra Rzeszotarska-Nowakiewicz, Tomasz Nowakiewicz, David de Sousa, Patrick Perin, Marzena Przybyła, Ulla Lund Hansen, Oleksii Sheremetiev, and Kirill Myzgin. I would also like to thank the editorial board since my intuition prompts me that this paper was the last arrival to this volume. 2 Actually, it is the background of any art-historian expertise to determine the authenticity of the work of art.

214 214 Maxim LEVADA demonstrates typical Roman stylistics with its roots deep in late Hellenistic tradition (fig. 1). Such belts occurred in barbarian princely graves, particularly in burial III in Wrocław-Zakrzów (Madyda- Legutko 1991: tabl. IV). Thanks to Mráv Zsolt s thorough study, the entire horizon of similar Roman ornaments was reanalysed to make its date younger; now it is usuwith the villa at the sea. This hoard was found by chance in municipality Kaiseraugust (Switzerland) in 1961 during a building. Unfortunately, the treasure was divided between different persons. Therefore, it is still not complete: the latest acquisitions of 1995 allows the one to draw the conclusion that some artefacts are still missing. The hoard was damaged by building machines and some of its ar- Fig. 1. Belt from Budapest-Pasarét (Hungarian National Muzeum) tefacts were being assembled from pieces for many years. So far the hoard consists of 270 silver artefacts, including 187 coins, its total weight is almost 60 kg (Kaufmann-Heinimann 2003). The Kaiseraugust treasure possibly belonged to a military officer or civil official of high rank, who served in a unit of the Roman army stationed in Castrum Rauracense (Kaiseraugst). It was buried at the foot of a wall in a solid wooden box. The hoard consisted of silver dishes, plates, spoons, jugs, cups, candelabrum, statuettes of Venus, three silver ingots with the stamp of emperor Magnentius, coins from Diocletian to Constantius II, 17 medallions of Constans, Constantius II, and Constantine II. This deposition might be interpreted as a part of gifts from Magnentius to the army honouring his ascension to throne in 350; judging by the fact that the latest coin was minted by Constantius II in 349, we have reasons to date the treasure to the period no later than the dethronement of Magnentius in 353 (Szidat 2003). The dish with fishes ( villa at the sea ) is decorated with the combination of three jewellery techniques in question along the horizontal edge and central medallion (fig. 3). The same techniques are used in design of the dish with an inscription in the medallion in honour of the tenth anniverally dated to the late third century. The mapping of these finds clearly points to the Limes on the Rhine and the Danube (Zsolt 2011: fig. 15). Roman silver finds decorated with a combination of nielloing, gilding, and by fine stamp or cutting tool disappeared in the end of the third or first half of the fourth century. There was a gap at least fifty years long, when the combination of such techniques was not used, by some reason. Moreover, silver artefacts with mercury-gilding became very rare. This gap is of great importance for the explanation of the fifth-century processes and the origins of the Sösdala style, so I will return to it below. If we continue the discussion of Roman artefacts at later period (when the Sösdala style appeared and became widespread), we are meeting an interesting fact that in the Late Roman period there were only a few silver artefacts combining nielloing, gilding, and decoration with fine stamp or cutting tool, definitely attributable to Roman manufacture. Most of them are ceremonial vessels: silver dishes with central medallions and decorated edge (fig. 2), of the first sub-type of Roman silver dishes according to Richard Hobbs (2010: 15, fig. 7). Primarily, it is a part of silver vessels from the Kaiseraugust treasure, particularly the famous dish

215 Sösdala: The Problem of Singling out an Artistic Style 215 Fig. 2. Silver dishes of Richard Hobbs s first sub-type showing central medallions and decorated edges with a combination of nielloing and gilding (not to scale). 1 Kaiseraugust treasure (Roman Museum of Augsburg), 2 Sevso treasure (Spencer Compton s collection), 3 Traprain Law, digital reconstruction (National Museum of Scotland) sary of Constans reign in 342/343: AVGVSTVS CONSTANS DAT LAETA DECENNIA VICTOR SPONDENS OM<I>NIBVS TER TRICENNALIA FAVSTIS, 3 and a portrait of a youth (Jeunesse Dorée) along the edge (fig. 4). Another example of similar Roman style is presented by some silver artefacts, also pottery, from the Sevso treasure. The history of this hoard is even more dramatic than of the Kaiseraugust treasure. The Sevso treasure appeared on London market in The legend stated that it originated from Lebanon. In 1990, an attempt was made to sell it at Sotheby s, and the scandal grew when governments of threes states, Yugoslavia, Hungary, and Lebanon, declared its illegal origin and put forward a claim that it belonged to them. At the moment, the treasure is owned by Spencer Compton, 7th Marquise of Northampton, and its further sales are blocked after the Hungarian government s appeal. The government of Lebanon considers that the finds were made in Tyre and Sidon. However, Yugoslavian (now Croatian) interpretation relates the find-spot with Barbariga on the southwestern coast of the Istrian peninsula. According to Hungarian version, the hoard was discovered in near Polgárdi east of Balaton lake. The latter seems the most convincing today. There is an account that the original hoard possibly consisted of as much as 30 artefacts (A Seuso-kincs 2012). The known part of the Sevso treasure consists of 14 silver vessels and one bronze cauldron. The chronology of the hoard has never been subject of a special study due to the circumstances explained above. Therefore, it is common to date it widely, within AD The upper date is well-grounded by an amphora with leopard handles, strikingly similar to the find in the famous grave discovered in 1812 г. at the Prut river, in Conţeşti, Romania (fig. 5). The biggest dish of the Sevso treasure forms a brilliant example of the technique under analysis, it might be produced in the same centre of jewellery as the dish from the Kaiseraugust treasure. It measures 70 cm in diameter and weighs about 9 kg. The dish shows an inscription placed along the medallion s edge: HEC SEVSO TIBI DURENT PER SAECULA MULTA POSTERIS UT PROSINT VASCULA DIGNA TUIS. That is the reason why the hoard is called Sevso (fig. 6). Another outstanding find including works of jewellery decorated with combination of gilding, nielloing, and fine stamp (and cutting tool) is the new hoard from Vinkovci in Croatia. It was discovered in 2012 by archaeological excavations in the downtown, which is an extraordinary case for the finds of the kind. The treasure consists of silver vessels, about 50 pieces total, weighing around 3 I am obliged to Nikita Khrapunov for consultation.

216 216 Maxim LEVADA Fig. 3. Central medallion on the dish with fishes from the Kaiseraugust treasure (Roman Museum of Augsburg) 36 kg. At the moment, the hoard is under restoration in the archaeological museum of Zagreb. Its preliminary chronology is the late fourth or early fifth century. The Vinkovci treasure includes several vessels with gilding and nielloing, particularly a dish with central scene that could be interpreted as Good Shepherd (fig. 7). Fragments of a dish with maritime scene on medallion and Jeunesse Dorée along the edge were in the famous hoard from Traprain Law (fig. 8). This complex is usually dated to (Curle 1920). The main contents of the hoard are luxury Roman vessels. The exceptions are a few demonstrative finds that should be mentioned. The first is a minor silver bi-plated brooch and a buckle with circular frame and rectangular panel, for ming single set (fig. 9. 2, 3). Similar brooches never occur in British islands, though they are typical of East Germanic antiquities in the northern Black Sea region and some of the neighbouring areas (Gavritukhin 2003). The decoration of three hemispheres on the foot-plate of the brooch from Traprain Law is similar to the find from the Baytal-Chapkan river on the North Caucasus (Gavritukhin 2003: fig. 3. 5).

217 Sösdala: The Problem of Singling out an Artistic Style 217 Fig. 4. Central medallion and fragment of edge decoration on the dish with inscription honouring the anniversary of Constans from the Kaiseraugust treasure (Roman Museum of Augsburg)

218 218 Maxim LEVADA Fig. 5. Amphorae from Conţeşti (State Hermitage Museum) and the Sevso treasure (Spencer Compton s collection) The second is a strap-end or a belt pendant (fig. 9. 1) of Markus Sommer s form C, type c (Sommer 1984: Taf. 22). Such ornaments are typical of the culture of Roman Germanic allies (phoideratoi) in the first half of the fifth century. And finally, the hoard includes fittings of luxury two-partite belt of silver with nielloing, gilding, and fine stamp decoration (fig. 9. 4, 5). A few parallels to this belt are known throughout Europe, however, without the combination of such techniques (Kazanski 1993; Nagy 2005: , fig. 27, 28; Levada 2011: 121, 134, fig , 15, 16). Silver belt fittings from grave 46 in the cemetery of Saint-Irénée (Lyon) in France are only nielloed (Des Burgondes: cat. no. 124). Nielloing on silver decorates fittings and buckles of the belt from Windisch-Oberburg (canton Aargau) in Switzerland and belt buckles from Tolna county in Hungary (Gallien: cat. no. 191; Fettich 1953: tab. LVII. 3 13). Bronze belt fittings from Luchistoye in the Crimea are engraved and decorated with fine stamp to imitate nielloing (Айбабин 2002). All these belts represent All-Germanic military culture, oriented toward the Empire, i. e. the same context as the belt-end under discussion. As for some other finds from the hoard, I will turn to them later, in the context of barbaric styles from the late fourth and fifth century. Back to Roman dishes, we should also discuss a unique find. Unfortunately, its tragic history is known only as fragments and it would never be known in full. It is the case of illegal discovery, probably in the Cherkassy oblast in the Ukraine. In November 2009, user with nickname lord1983 placed information at the Web site kladoiskateli.com about a hoard discovered three months earlier. As it comes from his/her words, there were several persons who found the treasure. They used metal scanner to discover 470 Roman denarii and several (seven?) fragments of a big dish. Total weight of the hoard was 1.5 kg. According to his description, the dish was decorated with nielloing and gilding, particularly an inscription translated by the finder as drink and have a long life. The finder also mentioned

219 Sösdala: The Problem of Singling out an Artistic Style 219 Fig. 6. Central medallion on a dish from the Sevso treasure (Spencer Compton s collection) Christogram on the dish. Several photographs published in the Internet allows us to see that the dish is of the same type as the finds in question (fig. 10, 11). In its centre are centaurs, though along the edge are a hunting scene, a composition with a flutist, and a small medallion with a head of youth (also Jeunesse Dorée). Two fragments show letters В 4 and Z. This dish is unique since the finder did not realize a feature which is well recognisable from photographs. What was called nielloing by him is actually sockets for blue enamel, which remains here and there. Therefore, we have enough reasons to think that similar enamelling existed on other silver artefacts from the period, though it did not survive. Probably, we need specific laboratory researches to confirm this interpretation. Current location of the dish is not known. According to an unconfirmed account, it was sold out to Moscow, where its track was lost The next find of Roman ceremonial vessels combining gilding, nielloing, and fine stamp with 4 BIBE ET?

220 220 Maxim LEVADA Fig. 7. Central medallion on a dish showing Good Shepherd from the Vinkovci treasure (Ministry of Culture of the Republic of Croatia) Fig. 8. Fragments of a dish from Traprain Law (National Museum of Scotland)

221 Sösdala: The Problem of Singling out an Artistic Style 221 Fig. 9. The finds in Traprain Law (National Museum of Scotland) cutting tool is the famous missorium of Kerch (fig. 12). This find was made by looters in 1891 in a secret compartment of a burial vault located in the Gordikovs family estate in Kerch, Crimea. It is considered that the missorium shows Constantius II, though Oleg Sharov has supposed here Constantine the Great (Шаров 2009). It is easy to notice that all the cited examples of Roman artefacts are ceremonyal vessels. The only exception is the belt from Traprain Law, which probably indicates Germanic phoideratoi. This environment possesses more examples of Roman jewellery art combining three techniques on the same artefact. However, they occurs first of all on belt fittings decorated in chip-carving (Kerbschnitt) style. Famous grave in Vermand contains excellent specimens of the same style (fig ). The assemblage is commonly dated to the turn of the fourth and fifth century (Schorsch 1986). Circular belt-ends with decoration of nielloing and gilding are known in Ejsbøl-Moor and Babenhausen (fig , 5; Sommer 1984: Taf , 3). A beltend and a fragment of belt fitting (?) were in the hoard of Traprain Law (fig , 4). Another treasure, at Coleraine in northern Ireland, included two belt fittings and a fragment of buckle (fig ). Similar belt fitting was discovered in Fosse, Norway (fig ). It is easy to notice that all these finds are of very similar style (fig. 13, 14). They are made in the same technique, so we can suppose a single centre of production or a workshop. However, from the point of view of the style, this is typical belt fittings of military culture of the Limes on the Rhine and the Upper Danube in the fourth and fifth century. Therefore, I cannot say that the attribution of some finds as works of Bernhard Salin s Style I is quite correct (see for example Myhre 2007). From the localization of almost all the mentioned finds outside of the main distribution area of fittings in chip-carving style, north of the Limes on the Upper Danube and Rhine cf. maps by Markus Sommer (1984), we have reasons to doubt that this centre of jewellery was located within the Empire. It would be even more difficult to discover Roman finds in plain style, i. e. without chipcarving. It is s surprise if we agree that the Sösdala style is related to Roman jewellery tradition. However, it is hard to find parallels to this unique find: there is one similar artifact, i. e.

222 222 Maxim LEVADA Fig. 10. A hoard from the Cherkassy (?) oblast

223 Sösdala: The Problem of Singling out an Artistic Style 223 Fig. 11. A hoard from the Cherkassy (?) oblast

224 224 Maxim LEVADA Fig. 12. The Kerch missorium (State Hermitage Museum) plate from illegal excavations somewhere in the Zaporozh ye oblast of Ukraine, which is known only from a photograph (fig ), and another plate with traces of fire discovered near Kucha village, Khmel nitskiy oblast (fig ). However, buckle frames from Czaszkowo are also decorated with triangles in niello, like the buckle frame from Vermand. Since this decoration is known on no other silver buckle, we have reasons to suppose that all these artefacts were produced in the same centre of jewellery. 5 No less luxurious is a belt set discovered in 1840 in Misery, very close to Vermand (fig , 12; Böhme 1974: 323, Taf. 128). Parallels to the Czaszkowo belt are the images of a griffin, a dog (?), and a lion. 5 In this regard, the combination of triangle patterns on the frame of buckle ( later? feature) and С-shaped border on its panel ( earlier? feature) at the Czaszkowo find should not embarrass. It is only an indicator of our imperfect knowledge of Roman military equipment from the second half of the third and first half of the fourth century because of a small number of finds after the collapse of the Limes in the third century and the lack of assemblages supplying reliable dates. As an example, similar border exists on definitely later clamp from the Coleraine hoard (see fig ).

225 Sösdala: The Problem of Singling out an Artistic Style 225 Fig. 13. Belt fittings in the chip-carving style with silver-gilding and nielloing. 1 4 Vermand (Metropolitan Museum of Art); 5 7 Coleraine Hoard (British Museum) Obviously, both the finds with chip-carving technique and plain artefacts represent mixed Romanbarbaric taste which influenced the entire style. An example of the same Roman style, or more precisely taste, from the same period could be belt-end from the cemetery of St-Marcel, made of gilt bronze with nielloing and silver encrustation. It shows Biblical scene with Daniel in the lions den (fig ). Now it is appropriate to return to the chronological gap in jewellery techniques under study. We should also remember another Scandinavian jewellery style from the Roman period, namely gold ornaments decorated with filigree and fine granulation (Andersson 1993a; 1993b; 1995). Ornaments in that style occurred in Scandinavia in the first half of the first century AD and disappeared in the beginning of the Late Roman period. However, in the third century these techniques became characteristic of Central Germanic antiquities, I mean the Zakrzów style. We can suppose with high degree of probability that the Scandinavian gold style appeared

226 226 Maxim LEVADA Fig. 14. Belt fittings in the chip-carving style with silver-gilding and nielloing. 1 Babenhausen (Sommer 1984; Land Museum of Mainz); 2 Fosse (Archaeological Muzeum of Stavanger; photo: Terje Tveit); 3 4 Traprain Law (National Museum of Scotland); 5 Ejsbøl-Moor (Ørsnes 1988; Sejrens triumph 2003) under the influence of outside contacts. Mark Shchukin has considered that this influence was related not to Rome, but rather to the first contacts between Sarmatians and Germans in the first century AD, so its direct prototypes were gold ornaments from Bosporos, Transcaucasia, or Eastern Mediterranean (Щукин 2005: 73 81). In my opinion, the idea behind these ornaments was borrowed in the Crimea. It was the place where gold lunula pendants with glass insets and granulation, gold bucket-shaped pendants with filigree decoration, and amber 8-shaped pendants, i. e. the ornaments typical of Germanic antiquities in the Late Roman period, occurred in the first century AD (Puzdrovskij, Zajcev 2004; Левада 2006: , рис. 9; Мордвинцева, Трейстер 2007: 94, 96 97). Therefore, late Hellenistic jewellery techniques came to the north of Europe in the first century, returned from there in the third century, but now not as isolated contacts but in result of migration of considerable groups of Germanic population. Similarly to the combination of nielloing, stamping, and gilding, a new gap appeared from the early fourth century when granulation and filigree came out of use. These techniques renovated only as late as the Hunnic period. Having paid attention to the total disappearance of granulation and filigree on gold artefacts from Scandinavia to Bosporos, Sharov has drawn a conclusion that seems paradoxical at the first look. The point is that a new, another style

227 Sösdala: The Problem of Singling out an Artistic Style 227 Fig. 15. Belt fittings in the chip-carving style with silver-gilding and nielloing. 1 8 Czaszkowo (photo: Aleksandra Rzeszotarska-Nowakiewicz, Tomasz Nowakiewicz); 9 Zaporozh ye (?) oblast ; 10 Kucha village, Khmel nitskiy oblast ; Misery (Alfred-Danicourt Museum)

228 228 Maxim LEVADA gradually developed from the mid-third century onwards. Bronze or silver artefacts were plated with gold foil with impressed decoration imitating the disappeared granulation, filigree, and even stamping. Such a new manner also intended many carnelian insets. The new fashion became especially distributed in the first half of the fourth century AD. What is most interesting, this new style initially appeared in barbaric environment, then it went over to Roman artefacts, the most exciting of which are helmets from Budapest and Berkasovo (Шаров 2010). Could both gaps showing the total decline of jewellery technology be a result of the same reason? I think they could. The key to the answer to this question is the fact that the gold-and-carnelian style from the early fourth century simply imitated earlier granulation and filigree. The reason is that the application of granulation and filigree design on surface is only possible by soldering with saturated solution of gold in mercury. The surface was prepared and covered with this solution, then the decoration of fine balls (granulation) or fine wires (filigree) is applied above. After that, the artefact was warmed so the mercury evaporated and the decoration is fastened tightly to the surface. However, the case of artefacts combining nielloing, stamping, and partial gilding also uses the same technique when gilt ornamentation was made with mercury solution. I held in my hands some of these artefacts, and careful visual examination allows to see strokes that sometimes step out of the zone of gilding outlined by cutting tool or stamp. This way, we can assert that both gaps in the development of jewellery techniques are caused by total disappearance of mercury. Its supplies probably declined or ceased as such in the second half of the third century. Quite likely, European barbarians and the Bosporan kingdom received mercury via the Imperial territory. It was actually the reason why the Empire started importing granulated and filigree ware later, since it could have some stocks of mercury or receive supplies at later. It is a natural question, then, where the so important mercury came from? I do not know specific studies of the problem in the Late Roman period, though the account of Pliny the Elder does not allow the one to answer the question. However, judging by abrupt decline of the jewellery skill, it is not the case of home production but of supplies from abroad. According to John F. Healy, the Empire obtained mercury from Iberia, Armani, and Ethiopia, i. e. from modern Spain, Transcaucasia ( from Armenians ), and East Africa (Healy 1978). Now we can remember that both eastern directions became problematic for Rome because of the growing power of the Sassanian realm, especially after 244, when Shapur I defeated them at Misiche, the more so after Valerian became captive at Edessa. The situation became catastrophic under Shapur II, who reigned from 309 to 379, i. e. in the period of the mercury gap. In 337, he broke the peace treaty of 297. Despite of the forced peace of 350, he started a new military campaign in 358, finishing with a profitable peace treaty of 363/64. Georgia and Armenia were among his most important acquisitions. Only after his death, under Shapur III, the peace of Nisibis was signed (384) to last for 36 years, and Armenia was divided between two empires. Certainly, the absence of specific study on the centres where mercury was mined and traded from the third to fifth century leaves my hypothesis as pure speculation. Nevertheless, I would like to remind that the best examples of ceremonial silver vessels decorated with gilding, nielloing, and stamp or cutting tool from that period originated from the Sassanian empire. Examples could be famous plate with a hunting scene from the tale of Bahram Gur Azadeh in the Metropolitan Museum of Art, leopard dish from the Miho Museum Kyoto, which is strikingly similar to Roman dishes with central medallion, and many others. It is indisputable that the plain style with gilding, nielloing, and stamping, synchronous with the finds in the Sösdala style, is almost not represented by Roman artefacts. On the other hand, the buckles under discussion are rather far from late antique aesthetics presenting a totally new, barbaric artistic tradition (fig. 16). Particularly, snakes on the buckles from Bar and Yalta (fig 16. 1, 3) were made by a Roman artisan in totally different way, according to the tradition rooted in the Late Hellenistic period. It is interesting to compare the Czaszkowo find (fig ) with fittings of the famous sword scabbard from Nydam (fig ). They have in common fanciful work and complicated design, including images of some monsters. Both these finds are pieces of jewellery of extraordinary high level, though presenting totally different manners. In both cases, we are dealing with initially wellplanned complicated composition showing different aesthetics not alike each other. It should be

229 Fig. 16. Belt fittings in the chip-carving style with silver-gilding and nielloing. 1 Bar Hoard; 2 Szabadbattyán (Hungarian National Museum); 3 Yalta (Yalta State Historical and Literary Museum); 4 Sagi (State Hermitage Museum); 5 Airan/Moult (Normandy Museum); 6 Bârlad Valea Seacă (Vasile Pârvan Museum)

230 230 Maxim LEVADA Fig. 17. Inscription on the backside of the buckles. 1 Szabadbattyán; 2 Bar Hoard; 3 Yalta added that the lion on the Czaszkowo fitting testifies that the artisan not only copied some sample but well understood which creature he was going to depict (fig ). The design of vine-leaf-cross on the buckles from Szabadbattyán and Airan/Moult could be a Roman, or more precisely Mediterranean, influence (fig , 5). However, the independent development of this style does not make impossible borrowings of decorative motives or jewellery techniques from the outside. The argument is chip-carving on the panel of the buckle from Bar (fig 16. 1). There is obvious problem with the establishment of the chronology of renovation of the jewellery techniques in study in the Empire. We cannot follow the method applied by Sharov for the case of the distribution of impression style using necessary number of reliable assemblages with Roman artefacts. There is only tentative chronology of the Czaszkowo find, if we agree that the coincidence of

231 Fig Saint-Marcel (Carnavalet Museum); 2 Nydam Mose (Sieg und Triumpf 2003); 3 Bar Hoard; 4 Tomb at Mesaksudi s land (Das Gold der Barbarenfürsten 2001); 5 Zdolbunov (?) district, Rovno oblast

232 232 Maxim LEVADA rare decoration on the buckle frame and analogous feature from Vermand is a chronological indicator. Therefore, we can suggest the date of ca. 400 AD, though it does not indicate that the Roman ware appeared earlier than Scandinavian. Moreover, the chronology of artefacts in the Sösdala style is developed at an advanced level by Jaroslav Tejral, since they indicate the Kachin Sösdala Untersiebenbrunn horizon. However, in the discussion of the early chronology of these barbarian artefacts, it is worth remembering another find, which has not been used in the context of the Sösdala style yet. Grave 507 from the Chernyakhov culture cemetery of Bârlad Valea Seacă contained a buckle, which is strikingly similar to the buckles from Bar, Yalta, Zamość, and Szabadbattyán, with the only difference in its small size (fig ). This find is extremely important because of two reasons. First, it forms the prototype for a series of bronze buckles of Renata Madyda- Legutko s type H43: Cecele 407, Podzamcze, Lublin, Strzegocice, Pyszków, and Gródek. Some of them have frames decorated with fine stamp, which makes them very similar to large buckles from the mid-fifth century of the type Strzegocice Tiszaladány Kerch (ML H44). Secondly, this rich grave has narrow chronology. Its terminus post quem is determined by a solidus of Constans II minted in Aquilea (September November 355; RIC 8: 332, no. 179), 6 re-shaped into a pendant. Igor Gavritukhin has paid especial attention to glass beaker from this grave dating it from the last quarter of the fourth to the first half of the fifth century and the complex in general to the late stage of the Chernyakhov culture (Гавритухин 2000: 269, рис ). The form of the comb and the second buckle from the grave allows us to infer, with a degree of uncertainty, that it is rather the late fourth century, so here we have the earliest buckle of the style in question. In order to understand who were the artisans making the ornaments in the Sösdala style, Romans or barbarians, we should pay attention to the finds from Bar, Kerch, and Szabadbattyán. As all know, on the back side of the buckle from Szabadbattyán is a Runic inscription readable as marŋ sd: North Germanic name Maring or Gothic Marings (fig ). The buckle from Bar has a pentagram scratched on the back side in the same manner (fig ). I had doubts for a little while if we had reasons to compare these facts or not, however, the help from Natal ya Turova (Yalta Historical and Literary Museum) allowed me to discover that the same pentagram occurred on the buckle from Yalta (fig 17. 3). As I already put it before, the buckles from Bar, Yalta, Szabadbattyán, and Airan/Moult were made by the same artisan (see Levada 2012). Now I think that we have much more reasons for such a conclusion. It comes out that, most probably, the inscription and signs on three buckles, discovered far from each other (Levada 2012: fig , 13, 15), were applied by the maker, rather than by the owners. Unfortunately, low panel of the buckles from Airan/Moult and Sagi do not survive. However, the buckle from Zamość, which is made in more rough way when the mentioned finds, does not have scratching. From the fact that the Runic inscription on the buckle from Szabadbattyán was written by the artisan, not by the owner, we have reasons to suppose that the artisan originated from the environment of Germanic barbarians. This conclusion, however, only refers to the cluster of finds which were the subject of my present analysis. As for the Sösdala style in general and the search for its origins, in my point of view the difficulty of this task is obvious, so the necessary condition is the making of a full catalogue of all the finds and their publication at a contemporary level. Therefore, I am would like to welcome again Charlotte Fabech s and Ulf Näsman s attempt to make a full publication of the Sösdala assemblage, the find which was much written about, though only a few persons probably saw it. Another find of extraordinary importance (yet illegal!), of the style very close to the buckles from Bar and Yalta, is a brooch discovered, possibly, somewhere in the Zdolbunov (?) district, Rovno oblast (fig ). The execution of design especially serpent s heads and the edging of semicircular heads of rivets is so similar to the design of the said buckles that here again we have reasons to suppose the making of these artefacts by the same artisan or in the same workshop. Therefore, we might reconsider the possibility of a production centre located somewhere between the Masłomęcz group area and the Southern Bug where prestigious works of jewellery were produced in a style brought from Scandinavia (Levada 2010: 78). 6 Attributed by Nikita Khrapunov.

233 Sösdala: The Problem of Singling out an Artistic Style 233 Appendix to the publication of the hoard from Bar (see Levada 2011) In the early 2013 I got informed about one more find from the hoard from Bar which was not known earlier. Unfortunately, all the attempts to learn about the localization of the hoard failed. The new find is a circular pendant with corrugated loop riveted to it, and through-hole in its centre. The edges are gilt, and the rest of its surface is decorated with nielloing in the form of a zigzag inscribed into a circle (fig ). When looking for parallels, I paid attention, due to Michel Kazanski s advice, to a find from the so-called tomb at Mesaksudi s land in Kerch. It was discovered by chance in the early twentieth century, therefore, we still do not understand if it was a single assemblage or a combination of artefacts from different periods. This grave contained, among others, a spatha with a very specific pommel. We are certainly dealing here with a re-shaped and re-used ornament. Original artefact was of silver with nielloing and gilding. It was circular, possibly with a central hole. On the edge are remains of a missing detail, possibly of a loop (fig ). This ornament undoubtedly belongs to the same circle, i. e. the Sösdala style. As is has already been stated, the secondary use remade this artefact as a pommel, pierced by a bronze nail with hemispherical gold-plated cap, which surface was decorated with gold granulation and almandines, typically of the polychrome style. This artefact (pendant?) was most likely re-made rather quickly, within the same chronological period. The said find is a good parallel to the pendant from Bar. Moreover, as I have already mentioned, the best parallels to narrow and long end-pieces of bridle straps from Bar occur in burial vaults at Kerch. All the mentioned facts make us to think why the bridles in the Sösdala style appeared in prestigious burial assemblages in various parts of Europe. We will probably be able to answer this question only after a thorough analysis of the Sösdala find, since the rider s equipment forms the greatest part of that unique deposition. Bibliography A Seuso-kincs és Pannónia: magyarországi tanulmányok a Seuso-kincsről I. The Sevso Treasure and Pannonia. Pécs, Andersson K. Romartida guldsmide i Norden. II. Fingerringar // Occasional Papers in Archaeology. 6. Uppsala, 1993а. Andersson K. Romartida guldsmide i Norden: Katalog // Aun. 17. Uppsala, 1993b. Andersson K. Romartida guldsmide i Norden. III. Övriga smycken, teknisk analys och verkstadsgrupper // Aun. 21. Uppsala, Bitner-Wróblewska A. From Samland to Rogaland. East- and West connections in the Baltic basin during the Early Migration Period. Warszawa, Böhme H. W. Germanische Grabfunde des 4. bis 5. Jahrhunderts zwischen unterer // Studien zur Chronologie und Bevölkerungsgeschichte. Münchner Beiträge Vor- und Frühgeschichte Curle A. O. Report on the excavation on Traprain Law in the summer of 1919 // Society of Antiquaries of Scotland Das Gold der Barbarenfürsten. Schätze aus Prunkgräbern des 5. Jahrhunderts n. Chr. zwischen Kaukasus und Gallien herausgegeben. Darmstadt, Des Burgondes à Bayard: mille ans de Moyen âge. Recherches Archéologiques et Historiques. Préface De Michel De Bouard. Catalogue d exposition. Grenoble; Lyon; Genève; Valence; Paris; Chambéry; Annecy; Bourgen-Bresse, Fabech Ch., Näsman U. A Lonely Rider? The Finding Place of the Sösdala Find and the Context of its Finds // Inter Ambo Maria. Northern Barbarians from Scandinavia towards the Black Sea (Abstracts of presentations). Simferopol, Fettich N. A Szeged-Nagyszéskósi hun fejedlemi sirlelet // Archeologia Hungarica XXIII. Gallien in der Spätantike. Von Kaiser Konstantin zu Frankenkönig Childerich. Ausstellungskatalog. Mainz, 1980.

234 234 Maxim LEVADA Gavritukhin I. Double-plate fibulas of subgroup I // A nyíregyházi Jósa András Múzeum Évkönyve. XLIV. Nyíregyháza, Healy J. F. Mining and Metallurgy in the Greek and Roman World. New York, Hobbs R. Platters in the Mildenhall treasure // Britannia Kaufmann-Heinimann A. Bilanz der Forschung seit 1984 // Der spätrömische Silberschatz von Kaiseraugst. Die neuen Funde: Silber im Spannungsfeld von Geschichte, Politik und Gesellschaft der Spätantike. August, Kazanski M. Les objets orientaux de I epoque des Grandes Migrations decouverts dans le couloir rhodanien // Antiquites nationales Levada M. To Europe via the Crimea: on possible migration routes of the northern people in the Great Migration period // Inter Ambo Maria. Contacts between Scandinavia and the Crimea in the Roman Period. Kristiansand; Simferopol, Levada M. Scandinavian Influence in the Crimea and on the Roman-Barbarian Borderland in the Fifth Century // International conference «Inter ambo maria. Contacts Between Scandinaviaand Crimea in the Roman Period» October, 21 25, 2010, Gaspra, Crimea, Ukraine. Simferopol, Madyda-Legutko R. Importe von metallenen Gürtelteilen des römischen Heeres im mitteleuropäischen Barbaricum // Archeologia XLII. Myhre B. Lye i Time på Jæren i Sørvest-Norge et glemt sentralsted fra eldre jernalder // [Internet: jenny-rita.org/bjomyh.html]. Nagy M., Zwei spätrömerzeitliche Waffengräber am Westrand der Canabae von Aquincum // AAH Oldenstein J. Zur Ausrüstung römischer Auxiliareinheiten: Studien zu Beschlägen und Zierat an der Ausrüstung der römischen Auxiliareinheiten des obergermanisch-raetischen Limesgebietes aus dem zweiten und dritten Jhdt n Chr. // BRGK Puzdrovskij A. E., Zajcev Ju. P. Prunkbestattungen des 1. Jhs. n. Chr. aus der Nekropole Ust -Al ma auf der Krim. Die Ausgrabungen des Jahres 1999 // Eurasia Antiqua Schorsch D. The Vermand Treasure: A Testimony to the Presence of the Sarmatians in the Western Roman Empire // Metropolitan Museum Journal Sieg und Triumpf Der Norden im Schatten des Römischen Reiches. Katalog zur großen Sonderausstellung in Kopenhagen Sejrens triumf, Norden i skyggen af det romerske imperium. København, Sommer M. Die Gürtel und Gürtelbeschläge des 4. und 5. Jahrhunderts im römischen Reich // Bonner Hefte zur Vorgeschichte Szidat J. Chronologische Übersicht der Jahre // Der spätrömische Silberschatz von Kaiseraugst. Die neuen Funde: Silber im Spannungsfeld von Geschichte, Politik und Gesellschaft der Spätantike August. Zsolt M. Egy késő 3. századi, niellóberakással díszített, középgyűrűs bronz szíjvégveret Brigetióból // Komárom- Esztergom Megyei Múzeumok Közleményei Zsolt M. Egy késő 3. századiniellóberakásos övlemez töredéke a Magyar Nemzeti Múzeumban // Corolla Museologica Kovács Tibor dedicata (Libelli Archaeologici Series. 4). Budapest, Ørsnes M. Ejsbøl I. Waffenopferfunde des Jahrh. nach Chr. // Nordiske Fortidsminder. Serie B. København Айбабин А. И. Поясной набор с пуансонным орнаментом // МАИЭТ IX. Гавритухин И. О. Финал традиций культур римского времени в Восточном Прикарпатье // Die spätrömische Kaiserzeit und die frühe Völkerwanderungszeit in Mittel- und Osteuropa. Łódż, Левада М. Е. «Другие германцы» в Северном Причерноморье позднего римского времени // БИ XI. Мордвинцева В. И., Трейстер М. Ю. Произведения торевтики в Северном Причерноморье 2 в. до н. э. 2 в. н. э. I. Древняя торевтика и ювелирное дело в Восточной Европе. 2. Симферополь; Бонн, Шаров О. В. Блюдо с изображением триумфа Императора из склепа Гордиковых в Керчи // Боспорские чтения. Х. Керчь, Шаров О. В. Парадная конская упряжь в эпоху поздней Империи // Краеугольный камень. Археология, история, искусство, культура России и сопредельных стран. II. М., Щукин М. Б. Готский путь (готы, Рим и черняховская культура). СПб., 2005.

235 Sösdala: The Problem of Singling out an Artistic Style 235 Максим ЛЕВАДА Сёсдала: проблема выделения художественного стиля Резюме Статья является продолжением дискуссии, возникшей на конференции «Inter Ambo Maria. Northern Barbarians from Scandinavia towards the Black Sea» в Гаспре (Крым) в октябре 2012 года после доклада Ш. Фабек и У. Несмана о знаменитой находке из Сёсдалы (Fabech, Näsman 2012). Авторы доклада не считают, что изделия в стиле «Сёсдала» производились северными варварскими мастерами (Fabech, Näsman 2012: 75). По их мнению, украшения этого круга являются римской работой, а изготавливались они на территории Империи. В этой статье речь идёт только об отдельной категории предметов этого стиля серебряных украшениях с декором, выполненным одновременно тремя ювелирными техниками: 1) мел ким штемпелем и (или) резцом, 2) чернью и 3) ртутной позолотой. Постановка проблемы выглядит следующим образом: необходимо максимально дистанцироваться от признаков хронологического горизонта «Качин Сёсдала Унтерзибенбрунн» и сосредоточиться исключительно на художественных особенностях стиля «Сёсдала», каким бы сложным не было их выделение. В своих рассуждениях я исхожу из того, что мастер-ремесленник высокого уровня а рассматриваемые изделия являются образцами высочайшей ювелирной техники работает только в одной, характерной ему ма нере. На этом основывается любая ис кусство ведческая экспертиза, определяющая под линность произведения искусства. Такая «кон сервативная» особенность ремесла и образует некие художественные стили и традиции. Изменение художественной манеры происходит либо на протяжении относительно длительного времени, либо под воздействием внешних глобальных импульсов. В последнем случае мастер изначально должен иметь образец для подражания, т. е. прототип. А для ху дожественного стиля «Сёсдала» мы таких стилистических образцов не знаем, следовательно, любые размышления о происхож дении, заимствовании или о внешних ре месленных центрах заходят в тупик. В данном случае не подразумевается заимствование технологических ювелирных при ёмов, а только попытка сравнить ху до жественные стили. Именно поэтому при вле кались синхронные находки украшений, в ко торых сочетаются обозначенные приёмы мелкий штемпель, чернь и ртутная позолота. Что касается всего стиля «Сёсдала» и поиска его истоков, то сложность этой задачи очевидна. Для её решения обязательным условием является полная каталогизация всех находок и издание их на современном уровне.

236 236 Mikhail LYUBICHEV The Secondary Used Artefact in Grave 41 of the Chernyakhov Cemetery of Voytenki Henceforth north-western refers to the elements of the Wielbark and Przeworsk cultures, as well as of other cultural groups of Central and Northern Europe, discovered in the sites of archaeological communities of the steppe and foreststeppe zones of Eastern Europe from the Roman period. The problem of the degree of their influence on cultural and historical processes in the latter region in general and their contribution to the genesis of the Chernyakhov culture in particular became one of the most important in the archaeology of Eastern Europe in the Later Roman period (Lyubichev 2012: 159). The north-western elements in the sites of the Chernyakhov culture including those in the area east of the Dnieper include the Chernyakhov / Wielbark kitchen and table ware in some of the settlements and in individual graves in certain cemeteries, wattle-and-daub structures discovered in the settlements, and bi-ritualism of the cemeteries (Магомедов 2001: 23, 33 34, 116). The Wielbark culture contains parallels to almost every type of the Chernyakhov cremations in pits and vessels (Магомедов 2001: 33 34). We can possibly relate the north-western element with the so-called complex type of cremation graves (Обломський 1999: 79), of Oleg Petrauskas types 3 and 4 (Петраускас 2002: 50 54). An indirect evidence of the north-western influence in this area comes from the finds of artefacts that played there absolutely another role in comparison with Central and Northern European Barbaricum. In this regard, especially interesting is a metal tubular pendant in grave 41 of the cemetery of Voytenki. 1 Its publication within the context of the grave seems an important task. The cemetery of Voytenki 1 is an inalienable part of the complex of archaeological monuments and sites of the same name, from the Late Roman and Early Migration periods, located in the forest-steppe area, on a side of a gorge with a small stream, near the sixth order river within the Dnieper water system (this river Mokryy Merchik Merchik Merla Vorskla Dnieper), at the watershed zone between the Dnieper and the Donets (Valki district, Kharkov oblast, Ukraine) (fig. 1. 1). It is under systematic investigations since 2004 by the Germanic-Slavonic Archaeological Expedition of the V. N. Ka razin Kharkov National University investigations uncovered 133 graves, including 68 cremations, 64 inhumations, and one cenotaph (fig. 1. 2) within this cemetery (Любичев 2009; 2011а; 2011b; Шультце, Любичев 2010; 2012). Grave 41 was probably located in the central area of the cemetery, because the site stretched from the north-west to the south-east, higher than the settlement, along the river (fig. 1. 2). The grave construction consists of a pit (undercut) and a step (shelf). The undercut was carved into the westerns side of the pit, and the entrance and step / shelf were located on the east (fig. 2). The pit measured 2.34 х 1.25 m at the top. The step / shelf had a vertical wall, with the only gentle slope on the south, near vessels and funeral food, from the step to the bottom of the undercut. Its surface was 1.16 m deep below the modern daily surface. The step / shelf was 2.24 m long, m wide, with maximum width at the south, and m high. The undercut / pit was 1.14 m long and m wide, m high, and was carved into the southern wall of the pit on 0.55 m and into the western wall on 1.14 m. Maximum depth of the undercut bottom is 1.3 m from the level of modern daily surface and 0.14 m from the surface of the step / shelf (fig. 2). Infill of the grave. The pit contained fragments of wheel-made vessels, bone comb, and silver fibula. The comb was located above the step / shelf, in its northern area, and the fibula was above the step 1 I am indebted to Maxim Levada who called my attention to the primary function of this artefact.

237 Fig location of the Chernyakhov culture, Illerup Ådal, and Voytenki; 2 plan of the cemetery of Voytenki ( excavations) with grave 41

238 238 Mikhail LYUBICHEV in its southern area, close to the vessels (fig. 4. 1). These goods were moved from the skeleton by earth-digging rodents. 1. Bone multipartite comb (fig. 4. 7, 12) has semicircular head and straight horizontal shoulders. Eight inset plates were fastened between the two plates with nine bronze rivets. The comb is 5.3 cm high, 9.8 cm wide, with its head 2.3 cm high and 6.3 cm wide at the base. The teeth are cm high. The comb belongs to Galina Nikitina s type IIIB1б (Никитина 2008: рис. 16). 2. Fine silver two-piece faceted brooch (fig. 4. 2, 14), with triangular sections of back-plate and footplate, and the cord in low position, on spring. It is 3.1 cm long, with the bow 0.35 cm wide and the spring 2.8 cm long. There are rectangular platforms on the head-plate, higher and lower than the fastener. The brooch is 3.1 cm long, with the bow 0.35 wide and the spring 2.8 cm long. The find belongs to Anatoliy Ambroz s series (Амброз 1966: 63 64) or Yevgeniy Gorokhovskiy s type Б1а (Гороховский 1988: 35). Skeleton of a young woman was extended to the north, declining at 330. The skull was slightly inclined, with the orbits to the east, the shoulders raised a little, the right upper extremity was along the body, with the palm down, and leaning against the wall of the undercut. Left upper extremity laid along the skeleton, slightly bent at the elbow and put behind the pelvis. Earth-digging rodents removed phalanges of the left of upper extremity, some of the ribs, and few vertebrae. The pelvis was wide, the lower extremities stretched, the fingers of the right lower extremity directed to the west, resting on the wall of the undercut. The left lower extremity was slightly bent at the knee (fig ). Total length of the skeleton was 1.48 m. Ceramic ware. The grave contained 11 ceramic vessels of various types (fig. 2. 2). The undercut / pit contained a bowl near the right elbow and an accumulation of four vessels on the south, near the funeral food (fig. 2. 2). 3. Wheel-made bowl, close form, carinated, with polished grey surface, flat base and the neck with a cordon (fig ). The vessel is 6.6 cm high, the diameter of the rim is 13.3 cm, of the body at the rib 13.7 cm, of the bottom 5,5 cm. Crushed by soil. The accumulation of vessels in the undercut consisted of three bowls and a pot. 4. Wheel-made bowl, close form, carinated, with polished grey and dark-grey surface, and flat base (fig ). The vessel is 9.7 cm high, the diameter of the rim is 21.5 cm, of the body at the rib 22.5 cm, of the bottom 8.2 cm. Crushed by soil. 5. Wheel-made bowl, close form, carinated, with polished grey surface, a cordon on the neck, and flat base (fig ). The vessel is 7.4 cm high, the diameter of the rim is 13 cm, of the body at the rib 14.6 cm, of the bottom 6.5 cm. The bowl slightly deformed under firing. 6. Wheel-made bowl, close form, carinated, with polished black surface (fig ). On the shoulder above the rib, there is a platform 1.3 cm wide, ornamented with zigzag line. The base is flat. It is 7.1 cm high, the diameter of the rim is 13.5 cm, of the body at the rib 15 cm, of the bottom 5.5 cm. 7 Wheel-made pot, with rounded sides, and coarse grey and dark-grey surface (fig ). On the shoulder of the vessel there are two horizontal parallel lines. The base is flat. The height is 11.8 cm, the diameter of the rim is 9.2 cm, of the body at the rib 11.8 cm, of the bottom 4.8 cm. On the step / shelf there were two accumulations of ceramic ware, no. 1 and no. 2. Accumulation no. 1 was located at the northern wall, and consisted of a vase / bowl with a small bowl inside (fig. 2. 2). 8. Wheel-made bowl / vase, close form, carinated, with polished grey surface, a cordon on the shoulder, and flat bottom (fig ). At the midheight of the body there is a belt of 27 grooves. The diameter of the rim is 26 cm, of the body 38.2 cm maximum, of the bottom 8.8 cm, the height 19.5 cm. Crushed by soil. 9. Wheel-made bowl, close form, carinated, with polished grey and greyish-brown surface, a cordon on the neck, and flat base (fig ). The diameter of the rim is 10.2 cm, of the body at the rib 11.3 cm, of the bottom 4.5 cm, the height 6.7 cm. Glass and ceramic cups inside vases forming a part of the set of vessels is a typical phenomenon in the Chernyakhov graves. Therefore, as well as because of the small size and profile of the bowl, we can suggest that it was used as a cup. Accumulation no. 2 was located closer to the southern wall of the pit. It contained two pots, bowl, and jug (fig. 2. 2). 10. Wheel-made double-handled jug with coarse surface of light-grey colour (fig ). The handles, of oval profile, are attached to the rim and shoulder of the vessels. On the lower neck there is a cordon. Three horizontal lines are scratched at the middle of the boy. The base is flat. The jug is 15.5 cm high, the diameter of the rim is 6.7 cm, of the body 14 cm, and of the bottom 8.3 cm.

239 Fig grave 41; 2 location of ceramic vessels; 3 sections of the grave

240 240 Mikhail LYUBICHEV 11. Wheel-made pot, with rounded sides, highly raised shoulder, and coarse light-grey surface (fig ). On the shoulder there is an ornamentation of two scratched parallel horizontal lines. The bottom is solid. A dark grey spot of soot on the body. The height is 15.2 cm, the diameter of the rim 13 cm, of the body 16.6 cm maximum, and of the bottom 7.7 cm. 12. Wheel-made bowl, open form, carinated, with flat bottom and grey and light-grey polished surface (fig ). The bowl is 5.7 high, the diameter of the rim 21.5 cm, of the body (at the carination) 19 cm, and of the bottom 6 cm. 13. Wheel-made pot, with rounded sides, maximum body diameter at the mid-height, and coarse surface of dark grey and brown colours (fig ). There are two parallel horizontal lines scratched on the shoulders. Black spots of soot on the surface. Low quality fabric with inclusion of large particles, thick walls ( cm) and flat base. The diameter of the rims is 10.8 cm, of the body 15 cm, of the bottom 6.1 cm, the height is 13.4 cm. Crushed by soil. Individual finds from the dead lady s costume and ornaments formed several accumulations. The accumulation at the upper skeleton combined the artefacts located below the skull, on the neck, thorax, near upper extremities, down to the waist inclusive (fig. 3. 1; 4. 1). At the neck bones and upper thorax there was a necklace of coral, carnelian and one glass beads (fig. 4. 8). There also were two matching fibulae: one at the left shoulder (fig. 4. 6) and another was removed by rodents to the east of the upper left extremity (fig ). A group of amber beads was located at the elbow of the right hand, close to the first bowl (fig ). A buckle was a bit higher than the waist (fig. 4. 5), clay spindle whorl was at the elbow of upper left extremity, on the inner side of it (fig. 4. 9). 14. Beads (seven specimens) of carnelian, lightbrown, faceted, prismatic, with cut corners, and truncated cone canal, measuring 0.6 х 1.0 to 1.5 х 1.1 cm (fig , 18 20) and belonging to type 1 of carnelian beads according to Oksana Gopkalo s classification (Гопкало 2008: 74). 15. Bead of carnelian, light-brown, faceted, prismatic, wit cut ribs and truncated cone canal, measuring 1.3 х 1 cm (fig , 21) and belonging to Gopkalo s type 3 of carnelian beads (Гопкало 2008: 75). 16. Bead of opaque black glass, of rounded ellipsoid bi-truncated form, with cylindrical canal (Fig ), measuring 1.0 х 0.4 cm (fig ) and belonging to Gopkalo s type II/14 (Гопкало 2008: 19). 17. Beads (97 specimens) of coral, white and light-pink, barrel-shaped, with cylindrical canal (fig , 23 30), belonging to Gopkalo s type 2 of coral beads (Гопкало 2008: 77). 18. Bronze two-piece fibula with returned foot, with cut decoration, segment cross-section of the back and foot, internal cord, and rectangular platforms on the head, higher and lower than the fastener. 4.7 cm long, with the bow 0.65 cm wide (fig. 4. 6, 15а). The spring is partly covered with remains of textile. It belongs to Ambroz s group 16, sub-group 2, series 1, variant 2 (Амброз 1966: 63 64) or Gorokhovskiy s type Б3б (Гороховский 1988: 35). 19. Bronze two-piece fibula with returned foot, cut decoration, segment cross-section of the back and foot, and rectangular platforms on the head, higher and lower than the fastener. The needle and the spring partly preserved. 4.7 cm long, with the bow 0.55 cm wide (fig , 15. b). It belongs to Ambroz s group 16, sub-group 2, series 1, vari ant 2 (Амброз 1966: 63 64) or Gorokhovskiy s type Б3б (Гороховский 1988: 35). Discovered near the skeleton, at the left upper extremity. 20. Beads (15 pieces) (fig ) of amber, cylindrical-bi-truncated-cone, with cylindrical canal, cm in diameter, belonging to Gopkalo s sub-group IV of amber beads (Гопкало 2008: 69). 21. Bronze buckle (fig. 4. 5, 13) with circular frame with widened front. The tongue is triangular in cross-section, with square platform at the junction to the frame, and the terminal overlapping the frame. The buckle from grave 41 has box-like panel, with the square top of the panel (box) and rectangular bottom. Leather strap (its remains are recorded in the grave) was fastened to the buckle with one bronze rivet. The frame measures 2.2 х 1.7 cm, with the tongue 1.9 х 0.35 cm, and the panel 1.2 х 1.5 cm and 1.5 х 1 cm. It belongs to Gorokhovskiy s buckle series Г (Гороховский 1988: 42). 22. Spindle whorl (fig. 4. 9), of clay, bi-conical, with concave bases and smoothed rib. Rough surface is grey, the diameter is 3.5 cm, the height 1.5 cm, the diameter of the hole 0.9 cm. Two pendants made of shells of marine molluscs and the tubular pendant pressed to each other and a bead located nearby formed the accumulation called below the pelvis and between the legs (fig )

241 The Secondary Used Artefact in Grave 41 of the Chernyakhov Cemetery of Voytenki 241 Fig. 3. Grave accumulation at the upper skeleton ; 2 accumulation below the pelvis and between the legs

242 242 Mikhail LYUBICHEV

243 The Secondary Used Artefact in Grave 41 of the Chernyakhov Cemetery of Voytenki 243 Fig. 5. Beads from grave amber; carnelian; 22 glass; coral Fig. 4. Grave location of finds; 12 comb; 13 buckle; 14 silver fibula; 15. a b two bronze fibulae; 16 pendant made of a shell of bolinus brandaris; 17 pendant made of a shell of cypraea pantherina

244 244 Mikhail LYUBICHEV 23. Pendant made of a shell of marine mollusc (bolinus brandaris) (fig. 4. 4, 16), with a hanging hole pierced through its upper part, measuring 6.6 х 5.2 cm. 24. Pendant made of a shell of marine mollusc (cypraea panterina) (fig. 4. 4, 17), with a hanging hole pierced through its upper part, with an iron ring passing through it (its remains stuck to the surface of the shell), measuring 6.6 х 4.2 cm. 25. Metal tubular pendant (fig. 4. 4; ) of cylindrical form is made of twisted bronze plate measuring 5.5 х 5.35 cm and 0.1 cm wide, covered with silver foil 0.02 cm thick. On either edge of the plate there are nine reach-through holes 0.2 cm in diameter. The tube made of the plate is 1.5 cm in diameter. At the top of the plate, there are remains of very corroded bow for hanging. In the tube, there is a fragment of textile cord, 5.4 cm long and 0.25 cm in diameter, twisted into a braid of four threads, each 0.05 cm thick (fig ). 26. Bead of olive opaque glass, rounded cylindrical, with cylindrical longitudinal canal. It is 1.4 cm long, 0.35 cm in diameter, with the hole 0.15 cm in diameter, belonging to Gopkalo s type VIII/1 (Гопкало 2008: 24). Funeral food. In the south-eastern corner of the undercut pit and on the gentle slope of the step there were remains of funeral food as two skeletons of a cock on the step / shelf, closer to the pit wall (1.12 m deep under the modern daily surface), and of a sheep, on the bottom of the pit / undercut and on the slope of the step (at the depth of m) (fig. 2. 1; 4. 1). Initial function of the tubular pendant and its place and role in the burial. The tubular pendant from the accumulation below the pelvis and between the legs is nothing but a twisted plate that initially had other function. In Illerup, such rectangular plates, some with straight sides and other with extensions with holes for rivets, formed details of sword belts (Carnap-Bornheim, Ilkjær, 1996а: Abb.46, 47, 49, 52, 53; 1996b: 22;1996c: Taf. 11, 25, 27, 28, 32, 33, 72, 74, 76, 77, 123, 125, 173, 174, 182, 184, 215, 217) (fig ). Similar plates also decorated horse straps (Carnap- Bornheim, Ilkjær 1996а: Abb. 39, 40; 1996b: 65; 1996c: Taf. 41, 42, 44, 101, 104, 110, 116, 160, 161, 195, 197) (fig ). In grave 41 of Voytenki, the pendant made of similar plate forms a part of a single amulet with pendants made of shells of marine molluscs cypraea and bolinus (fig. 3. 2; 4. 4). Amulets combining shells of these marine molluscs were widespread in Germania Libera from the third to the seventh century (Schulze-Dörrlam 1986: ). These amulets are also well-known in the Chernyakhov area that formed a component part of Germania Libera (Гопкало 2008: карта 36). Shell pendants appear in different combinations, for example one cypraea and two bolinus shells in grave 337 in the cemetery of Danceni (Рафалович 1986: табл. L), or two cypraea shells in grave 224 in the same cemetery (Рафалович 1986: табл. XXXIX). In the Chernyakhov culture, specific women s ornamental amulets combined shell pendants with various artefacts (Гопкало 2011: 142): need le-case and bone in grave 5 in the cemetery of Sumy Sad (Некрасова 1985: рис.4), or fang pendants in grave 25 in the cemetery of Furmanovka (Сымонович 1988: рис.14). Grave 111 in the cemetery of Voytenki contained pendants of shells cypraea and bolinus accompanied with bone needle-case and an ornament of a sawn-off end of deer s antler (Любичев, Мызгин, Варачева 2010: 36). There also is a combination of shell pendants and prismatic bone pendants, the so-called Heracles maces (Herkuleskeulе), with a certain regulation when the inhumation of a woman with the head to the north is accompanied with belt amulet of shells and Herkuleskeulе (Бобровська 1999: 89, 91). In grave 21 in Koblevo two shell pendants (cypraea, bolinus) were accompanied with a needle-case, a needle, and two Herkuleskeulе artefacts (Сымонович 1979: 78, 81, рис ; ). In grave 132 in Velikaya Bugayovka there are accumulation of cypraea shell, Herkuleskeulе and a bead, and another one with bone needle- Fig the pendant from burial 41 in Voytenki; 3 4 the cord from the pendant; 5 7 horse harness fitting from Illerup Ådal (Carnap-Bornheim, Ilkjær 1996c: Taf. 77: WUL; Carnap-Bornheim, Ilkjær 1996a: Abb. 53: WUM; 47); 8 11 sword belt fitting from Illerup Ådal (Carnap-Bornheim, Ilkjær 1996c: Taf HGP; Carnap-Bornheim, Ilkjær 1996a: Abb. 39. RWM; 40)

245 The Secondary Used Artefact in Grave 41 of the Chernyakhov Cemetery of Voytenki 245

246 246 Mikhail LYUBICHEV case and a piece of labdanum (Петраускас 2004: рис ). Such combinations are typical for the Chernyakhov culture and also for other parts of the Germanic world, particularly to Frankish, Alemannic, and Bavarian graves from the fifth to the seventh century (Werner 1964: ). This amulet is interpreted as a fertility amulet protecting the marriage of a man and a woman from evil forces and helping them to have children (Werner 1964: 181). Amulets of sea shells are considered an attribute of women of a certain age (Lennartz 2004: 165). The combination of the mentioned artefacts in this amulet obviously symbolises the joining of the sex organs of a man and a woman leading to a child-bearing (fertility). Therefore, these amulets are typical of women s graves. It is interesting with grave 41 of Voytenki that bone prismatic pendant or needle-case was replaced there by a tube of metal plate. It is obviously a phallic symbol. The phallic pendant itself in combination with two bolinus sells was in child s grave 1 at castellum Celesium / Pförring in the Danube area (Schulze-Dörrlam 1986: 350, Abb. 5). The initial function of the metal pendant from Voytenki was also related to male sub-culture being a part of warrior s and horseman s equipment. However, we can not be sure that the amulet of marine shells combined with prismatic pendants, needle-cases, and tubes had the only, mentioned above, symbolism and were present in graves of young women only. The above-described grave 5 in the cemetery of Sumy Sad belonged to an elderly woman (Некрасова 1985: 76 78, рис. 4). Was it the case of parous and nonparous women? Possibly, these amulets were connected with nonparous women of different age. Only anthropological research can find an answer to this question. The ethnic and cultural attribution of the grave. The grave consists of a pit with an undercut, doubtless a Sarmatian / Late Scythian feature (Магомедов 1999: ; Гей : ; Любичев 2011: and табл. 1). In the cemetery, around grave 41, there also are burials in pits with undercut (fig. 1. 2). Grave 41 contained 11 ceramic vessels. Big number of vessels in grave could also be considered a Sarmatian / Late Scythian feature (Седов 1978: ; Kokowski 2004: 21 24, 29, 35, 37). Around grave 41 there are assemblages with a great number of vessels (fig. 1. 2) (their quantity in graves no. 86/1, 96, and 101 is even bigger, 16 pieces in each). Funeral food remains as traces of complete skeletons of animals or individual bones, which is also considered a possible Sarmatian / Late Scythian feature (Рикман 1975: 319; Седов 1978: ; Магомедов 1999: ; Гей : ). Graves with this feature are known in the cemetery of Voytenki (fig. 1. 2). Although meat products also appear in Germanic burials (Schulz 1953: 13), it looks like a singular phenomenon typical of rich graves in Germanic world of Central Europe. Some researchers consider numerous big and fine beads another Sarmatian / Late Scythian element (Рикман 1975: 319; Гей : ; Kokowski 2004: 23 24). The same is the interpretation of pendants of shells of marine molluscs (Diaconu 1966: ; Kokowski 2004: 21), however, as it has been discussed already, these pendants in combination with bone prismatic amulets are similarly typical of the Germanic world of Western and Central Europe. Two fibulae, a comb, amulets Herkuleskeulе, amber beads, and cypraea shells are considered elements of East Germanic tradition (Петраускас 2004: 181). In the given case, we have the third, small fibula. The combination of two matching fibulae with returned foot of medium size located near shoulders and a small fibula below them is not typical of the Chernyakhov women s costume (Гопкало 2011b: 28 29, рис. 2). These features altogether within the same grave suggest that the burial combined the Late Scythian / Sarmatian and Germanic (north-western) features. The chronology of the grave. The date of grave 41 is determined by chronological indicators of the first level (fibulae, buckle, and comb) accompanied with chronological indicators of the second level (ceramic ware, beads, and spindle whorl) with wider dating. The chronology of Voytenki cemetery is based on a seriation, which discovered two groups of burials in the cemetery (Шультце, Любичев 2012: 414). Grave 41 belongs to the second group, particularly typical of fibulae with returned foot Gorokhovskiy Б3б and Б1а, buckles Gorokhovskiy Г, and bone combs Nikitina IIIB1б (Шультце, Любичев 2012: ). This group correlates with Gorokhovskiy s stage 5 of the Chernyakhov cemeteries in the Ukrainian forest-steppe area (Гороховский, 1988: 44 45) and Jaroslav Tejral s stage D1 (Tejral 1986: ), i. e. the last quarter of the fourth and early fifth century as absolute dates. This grave dates back to the last stage of the Chernyakhov

247 The Secondary Used Artefact in Grave 41 of the Chernyakhov Cemetery of Voytenki 247 culture also after the box-like design of the outer panel of the buckle, typical of the Hiunnc period (Петраускас, Шишкин 2009: 224). Oksana Gopkalo has attributed the Chernyakhov graves with shell amulets to Gorokhovskiy s stage 4 or Jaroslav Tejral s phase С3/D1, interpreting the appearance of the great deal of such goods as a result of trading in AD 330/ (Бобровська 1999: 90 91). Quite probably, shell pendants circulated among the Chernyakhov population even when that trade finished, in the last quarter of the fourth or in early fifth century. Bibliography Carnap-Bornheim C. von, Ilkjær J. Illerup Ådal. 5. Die Prachtausrüstungen. Тextband. Aarhus, 1996a. Carnap-Bornheim C. von, Ilkjær J. Illerup Ådal. 6. Die Prachtausrüstungen. Katalog, Fundlisten und Litera tur. Aarhus, 1996b. Carnap-Bornheim C. von, Ilkjær J. Illerup Ådal. 7. Die Prachtausrüstungen. Tafelband. Aarhus, 1996c. Diaconu Gh. Spätsarmatische Elemente in der Sîntana de Mureş-Tschernjachow-Kultur // Dacia X. Kokowski A. Przyczynek do historii kontaktów germanów i sarmatów w świetle badań archeologicznych // Sarmaci i Germanie. Lublin, Lennartz A. Die Meerschnecke Cyprea als Amulett in Frühen Mittelalter. Eine Neubewertung // Bonner Jahrbücher Bd Lyubichev M. North-Western Elements of the Sites from the Late Roman Period in the Forest-Steppe Area between the Dnieper und the Donets // Inter Ambo Maria. Northern Barbarians from Scandinavia towards the Black Sea (Abstracts). Simferopol, Schulze-Dörrlam M. Gotische Amulette des 4. und 5. Jahrhunderts n. Chr. // Archäologisches Korrespondenz blatt Schulz W. Leuna. Ein germanischer Bestattungsplatz der spätrömischen Kaiserzeit. Berlin, Tejral J. Fremde Einflüsse und kulturelle Veränderungen nördlich der mittleren Donau zu Beginn der Völkerwanderungszeit // Archäologia Baltica VII. Werner J. Herkuleskeule und Donar-Amulett // Jahrbuch des RGZM Амброз А. К. Фибулы юга Европейской части СССР / САИ. Д М., Бобровська О. В. Жіночі поясні амулети черняхівської культури // Археологія Гей О. А. Проблема выделения скифо-сарматских элементов в черняховской культуре // Stratum plus Гопкало О. В. Бусы и подвески черняховской культуры. Киев, Гопкало О. В. К реконструкции черняховского убора и костюма // Археологія і давня історія України. 5. Археологія: від джерел до реконструкції. Київ, 2011а. Гопкало О. В.Чоловіче та жіноче черняхівське вбрання // Археологія. 2011b. 2. Гороховский Е. Л. Хронология черняховских могильников Лесостепной Украины // Труды V Международного конгресса археологов-славистов. Т. 4. Киев, Любичев М. В. Погребение с римским импортом на могильнике черняховской культуры Войтенки (Восточная Украина) // Ostrogothica. Археология Центральной и Восточной Европы позднеримского времени и Эпохи Великого переселения народов. Харьков, Любичев М. В., Мызгин К. В., Варачева К. Г. Отчёт о работе Германо-Славянской археологической экспедиции Харьковского национального университета имени В. Н. Каразина в 2009 г. Харьков, 2010 (The Scholarly Archive at the Institute of Archaeology of the National Academy of Sciences of the Ukraine. File 2009/147). Любичев М. В. О поздней дате черняховской культуры к востоку от Днепра (комплексы ступени D1 могильника Войтенки) // Древности Харьков, 2011а. Любичев М. В. Погребения с позднескифскими / сарматскими признаками на могильниках черняховской культуры в днепро-донецкой лесостепи // Oium. Київ, 2011b. Магомедов Б. В. Сармати у складі черняхівської культури // Етнокультурні процеси в південно-східній Європі в I тис. н. е. Київ, Магомедов Б. В. Черняховская культура. Проблема этноса. Lublin, 2001.

248 248 Mikhail LYUBICHEV Некрасова Г. М. Охоронні розкопки черняхівського могильника поблизу м. Суми // Археологія Никитина Г. Ф. Черняховская культура Поднестровья (по результатам анализа археологических источников). М., Обломський А. М. Типи поховань на черняхівському могильнику Компанійці (етнокультурна інтерпрета ція) // Археологія Петраускас О. В. Типи археологічних комплексів трупоспалень черняхівської культури (територія поширення, етнокультурні особливості та хронологія) // Археологія Петраускас О. В., Шишкин Р. Г. Некоторые результаты исследования поселения черняховской культуры Великая Бугаевка-1 // Vita antiqua Петраускас О. В. Поховання багатої дівчинки на могильнику черняхівської культури Велика Бугаївка // Археологія давніх слов ян. Київ, Рафалович И. А. Данчены. Могильник черняховской культуры III IV вв. н. э. Кишинев, Рикман Э. А. Этническая история населения Поднестровья и прилегающего Подунавья в первых веках на шей эры. М., Седов В. В. Скифо-сарматские элементы в погребальном обряде черняховской культуры // Вопросы древней и средневековой археологии Восточной Европы. М., Сымонович Э. А. Коблевский и Ранжевский могильники около г. Одесса // МЧК. М., Сымонович Э. А. Придунайский могильник Фурмановка // МЧК. М., Шультце Э., Любичев М. Керамика из могильника и поселения Войтенки 1 (Восточная Украина): сравнительный анализ // Лесная и лесостепная зоны Восточной Европы в эпохи римских влияний и Великого переселения народов. Ч. 2. Тула, Шультце Э., Любичев М. Хронология могильника черняховской культуры Войтенки 1 (Восточная Украина) (по материалам раскопок гг.) // Лесная и лесостепная зоны Восточной Европы в эпохи римских влияний и Великого переселения народов. Тула, Михаил ЛЮБИЧЕВ Вторичное использование предмета в погребении 41 могильника черняховской культуры Войтенки Резюме В погребении 41 (яма с подбоем) могильни ка Войтенки обнаружен амулет, включавший подвески-раковины bolinus и cypraea, а также металлическую подвеску в виде трубки. Трубка была скручена из бронзовой плас тины с серебряным покрытием, в ней находи лась нить. Подобные пластины в Иллерупе являются частью портупеи меча и конской сбруи. В этом амулете трубка как бы заменяет костяной игольник или призматическую костяную подвеску (Herkuleskeulе). В погребе - нии сочетаются позднескифские / сарматские и северо-западные (германские) черты. Оно относится к ступени D1 по Я. Тейралу и фазе 5 по Е. Л. Гороховскому (конец IV начало V вв.).

249 249 Magdalena MĄCZYŃSKA A Gold Brooch from Młoteczno (Hammersorf), Braniewo Powiat, in North-Eastern Poland 1 The Młoteczno finds, discovered between 1873 and 1917, gradually gave birth to numerous interpretations, particularly those discussing their homogeneity as a single complex, chronology, and origin. There are at least 17 gold and silver artefacts unearthed in various areas of the village of Młoteczno, of which we now know medallion of Constantius II, produced in Constantinople in AD (Bursche 1998: 237), three gold bucketshaped pendants, polychrome gold brooch with a chain, fragments of two silver eastern Roman vessels (lances), two gold neck-rings, and possibly an urn with cremated bones. Most of these finds appeared in the collection of the Prussia Museum in Kaliningrad (Königsberg, Królewiec), one pendant was melted down, and two gold neck-rings, which had been in the Ethnological Museum, Berlin, before 1945, were discovered in the Pushkin Museum in Moscow (Merowingerzeit 2007: Kat. IV ). Other artefacts should be considered missing. The first find of 1873 consisted of fragments of two silver vessels (Hirschfeld ): a patera with hunting scenes, of Byzantine origin, and a plate with flora design, a Gallic or Italian product. Both vessels cannot be later than the early fifth century (Bott 1982: 151 f.; Munksgaard 1987). Gold brooch was discovered in 1913, and three pendants and two neck-rings, making the main concentration of the finds, and a medallion from more than two kilometres to the east of them, appeared four years after. Fragments of silver vessels were discovered in between of these two locations. Felix Ernst Peiser, who was the first to write about all these finds (Peiser 1919), and Nils Åberg after him (Åberg 1919: 67 f.) have interpreted this collection as a single assemblage originating from a barrow grave from the mid-fifth century (Peiser) or from ca. 400, the first half of the fifth century as the latest (Åberg). Max Ebert has suggested another interpretation (Ebert 1923: 159) following his studies of the finds and related documents de visu in the museum in Konigsberg. He has divided the artefacts into early and late groups, the former including the brooch, three pendants, and silver vessels, dated by him to the second half of the fourth century, or ca. AD 400 as the latest. The late group consists of two neckring of Scandinavian type from the first half of the sixth century, most likely from ca. 550 (Ebert 1923: 170 ff.). Günter Mangelsdorf s recent research has suggested the sixth century as the general date of the latter (Mangelsdorf 2011: 97, Kat. 6). Many years after the discovery, Aleksander Bursche has reconsidered the location of the Młoteczno finds (Bursche 1998; 76 79). Against the background of the visual survey of the place and field research, he has run to the conclusion that these finds were perhaps votive offerings originated from adjacent turf accumulations, which had been discovered as early as Ebert s times (Ebert 1923: 156) (similar to: Cieśliński, Nowakowski 2005: 262). This interpretation seems the most probable if one considers 200-year-long chronological gap between different artefacts. Although Volker Hilberg has run to a similar conclusion (Hilberg 2009: 161f.), he has analysed the location of three find spots to learn that they were located in the uppermost places in this area, which could not be turf mines. The gold polychrome brooch has been subject of repeated studies and references in scholarship. The latest researcher able to see it with his own eyes was Ebert (1923: ). The brooch measured 13.2 cm in length, weighing 92.9 g (initially ca. 100 g). Its semicircular head-plate had three knobs in the form of stylized beast s heads. The headplate, bow, and diamond foot-plate were decorated with insets of semiprecious stones, filigree, and granulation. Unfortunately, none of the previous publications attributed the stones, when it was still possible. Only Peiser writes that, besides semiprecious stones, the brooch was decorated with amber insets, not mentioned by Ebert (Peiser 1919: 93). In 1 The preparation of this paper has been funded by the National Science Centre of Poland as grant DEC- 2011/02/A/HS3/00389 The Great Migration in the basin of the Oder and the Vistula.

250 250 Magdalena MĄCZYŃSKA Fig. 1. Localization of the finds in Młoteczno: 1 coins (modern); 2 preserved pendant; 3 melted pendant; 4 necklaces; 5 brooch; 6 fragments of silver vessels; 7 golden medallion (Hilberg 2009) the archive of the Museum of Prehistory and Early History, Berlin, there is a clipping labelled PM-A 1562/1, without newspaper title, that informs about garnet insets. 2 Besides, on the head-plate there were stylized S-shaped sea-monsters made with filigree and granulation, and a fragment of gold chain with 29 links, weighing 16.2 g, similar to the chain of polychrome brooches from Airan / Moult (Salin, France, Lanord 1949; Pilet 2007) survived near the head-plate. On the opposite side of that chain there probably was an identical brooch. The combination of filigree, granulation, and polychrome style of the Młoteczno brooch makes it considered in a few years after discovery to be the intermediate form between the brooches of type Zakrzów, polychrome artefacts in Untersiebenbrunn style, and early variants from the hoard of Szilágysomlyó. This interpretation, following initial remarks by Peiser (1919: 95) and Nils Åberg (1919: 67 ff.), was formed by Ebert (1923: 167), who suggested its date as the second half or end of the fourth century, about AD 400 at the latest. Later on, this concept was revived by other researchers, e. g. Hans Bott (1982: 142 ff.). Nandor Fettich has paid attention to the seamonster design of the brooch from Młoteczno and 2 I am indebted to professor Aleksander Bursche for this information.

251 A Gold Brooch from Młoteczno (Hammersorf), Braniewo Powiat, in North-Eastern Poland 251 Fig. 2. Młoteczno (Hirschfeld 1886) two more clasps from Szilágysomlyó II (Fettich 1932: pair of brooches no. VI, pl ; 17. 2; fig. 4; Kiss 1999: Kat ), thus drawing a line of development from the buckle with carnelian inset in grave 2 in Wrocław-Zakrzow, via the brooch with filigree and carnelian insets in Årslev on Funen (Storgaard 1994), to the brooch from Młoteczno. Its decorative monsters could be borrowed from the Mediterranean environment (Fettich 1932: 64 66). Mogens Mackeprang has also supposed Hunga rian origin of the Årslev brooch (Mackeprang 1940: 89). A very similar motif with seahorse (hippocampus antiquorum) is known on silver gilt buckle from Sagi in vicinity of Kherson at the estuary of the Dnieper (Annibaldi, Werner 1963: 371 ff.; fig a; pl ), now in the State Hermitage collection. 3 It was discovered in 1887 together with silver fittings of horse harness and gold appliqués featuring granulation and cloisonné. This find dates back to the age of Attila, and the buckle decoration of a mask encircled with two snakes and stylised fish, according to Joachim Werner, should be derived from late antique Danubian environment rather than the Black Sea area. Maxim Levada holds another opinion (Levada 2011: 120, рис ) relating the buckle of interest with the bog find from Finnestorp, which is 3 I am indebted for this picture to Maxim Levada and Vyacheslav Baranov (Kiev). The buckle from Sagi (similar to the find from Alyoshki) has been published by Aleksandr Aibabin (Айбабин 1990: 29, рис ), though Irina Zasetskaya calls the place itself, without a reference to this very buckle, Alyoshki / Kuchugury (Засецкая 1994: табл. 15). The find condition and all the grave goods were recently published by Baranov, who interpreted them as a destroyed cremation grave from phase D2/D3 (Баранов 2010).

252 252 Magdalena MĄCZYŃSKA Fig. 3. Młoteczno (Merowingerzeit 2007)

253 A Gold Brooch from Młoteczno (Hammersorf), Braniewo Powiat, in North-Eastern Poland 253 Fig. 4. Młoteczno (Peiser 1919)

254 254 Magdalena MĄCZYŃSKA Fig. 5. Młoteczno (Peiser 1919) Fig. 6. Szilágysomlyó (Kiss 1999)

255 A Gold Brooch from Młoteczno (Hammersorf), Braniewo Powiat, in North-Eastern Poland 255 Fig Sagi (Levada 2011); 2 Gudme (Østergaard Sørensen 1994) silver gilt and also with human mask (Nordquist 2007). The frame of the latter buckle is formed by two snakes. Having analysed the hoard of metal ware from Bar in the Ukraine, which lays within the Untersiebenbrunn Kachin horizon and contains artefacts in Sösdala style, Levada has supposed that hoard to reflect migration of some Scandinavian elites towards the Black Sea and the Danube. Therefore, the Sagi buckle is probably of Scandinavian origin as well (Levada 2011: 119 n.; 134). Kazimierz Godłowski (1981: 79 ff.) has attributed the Młoteczno brooch to the polychrome finds of type Nezhin Szilágysomlyó Regöly Untersiebenbrunn (Tejral 1988: ; Tejral 1997: ; Bierbrauer 1995: ), though hollow filigree endings of its front bow and axle bar possess features typical of ornamented brooches from Pomerania, such as Kiełpin, Świelino, and Trzebiatów (La Baume 1925: 20 27; Zeiss 1937; Eggers, Stary 2001: pl ; ), thus suggesting its local origin. However, similar endings of axle bar and head-plate area also known with the brooches from grave in Úherce in the Czech Republic and with brooches of type Wiesbaden from the hoard in Świlcza, Rzeszów powiat (district) in the south-eastern Poland (Svoboda 1965; Werner 1981: 241 ff.). Werner considered Pomeranian and Scandinavian hollow head-plate endings to be a local modification of beast s heads of Danubian brooches (Werner 1981: 242). Radu Harhoiu has compared 29 polychrome brooches to place the Młoteczno find to group III with brooches from Gelénes and Szilágysomlyó (Harhoiu ; Harhoiu 1998: 93 97, pl. 99); the latter artefacts have a big carnelian inset on diamond foot-plate instead of seahorses (Kiss 1999: Kat ). Group III occupies an intermediate position between the Dnieper and Kerch brooches of group II from the third quarter of the fourth century and those from Velţ, Airan, Untersiebenbrunn, and Rábapordány of group IV from the early fifth century. The researcher has dated group III, the Młoteczno brooch in particular, to the late fourth century, with a reference to the eyelet on the Młoteczno medallion with analogies in the hoard from Szilágysomlyó (Harhoiu 1998: 96). However, Harhoiu has included the finds from Nezhin, Porshnino, Regöly, and Szilágysomlyó with beast s heads (see e. g.: Кропоткин 1970:

256 256 Magdalena MĄCZYŃSKA Fig Nezhin; 2 Untersiebenbrunn; 3 Rábapordány; 4 Regöly (Merowingerzeit 2007; Germanen 1987)

257 A Gold Brooch from Młoteczno (Hammersorf), Braniewo Powiat, in North-Eastern Poland 257 Fig. 9: 1 4 Świelino; 5 Kiełpino (La Baume 1925; Zeiss 1937) кат , 1044; рис , 7; 54; Godłowski 1981: 79 ff.), similar to the Młoteczno brooch and often referred to in its context, into his group V from the second half of the fifth century. Michael Schmauder has interpreted the Młoteczno brooch as similar to the finds from Regöly, Rábapordány, and the Metropolitan Museum, New York (Schmauder 2002: Kat. XV ; XVI; XX ; pl ; 115; 116; ). The latter one most likely originates from the hoard in Szilágysomlyó (Bierbrauer 2000). Granulation on the ends of two very simple brooches from Rábapordány looks similar to the technique applied to the Młoteczno brooch, however, it is necessary to underline the Młoteczno brooch s higher artistic level.

258 Fig. 10. Szilágysomlyó (Dembski 1999) Fig. 11. Zamość (Kokowski 1995)

259 Fig. 12: 1 3 Wrocław-Zakrzów; 4 Stráže; 5, 8 Hungary, unprovenanced; 6, 7 Ostrovány (Quast 2009; Hampel 1905; Prohászka 2006)

260 260 Magdalena MĄCZYŃSKA Fig. 13. Sanderumgård (Jørgensen, Petersen 1998) Robert Stark has reconsidered the origin of the Młoteczno brooch when discussing the brooches from Szilágysomlyó, especially those with the seahorse design. (Stark 1999: 142, 149, 154). He has placed it in his group I, presumably of eastern Roman artefacts, from the late fourth and early fifth centuries, interpreting the Młoteczno brooch as a local artefact and supposing that the artisan who made it did not had enough polished stones and therefore used granulation and filigree, the techniques he knew perfectly well, which were widespread in the lower Danube and northern Black Sea areas. Hilberg also leans towards that interpretation (Hilberg 2009: 159 ff., fig ), underlining Scandinavian features of the brooch, particularly filigree and granulation, and interpreting hollow knobs on its head-plate as local attributes. In recent years, Adam Cieśliński has paid much attention to the Młoteczno find (Cieśliński 2008: ; Cieśliński 2010: ). Having discussed the origin of the brooch, he acknowledged that granulation, twisting, and filigree on wires composing the sea-monsters indicate its local origin in contrast to embossing used in brooches from Szilágysomlyó with similar ornamentation, and the samples inspiring the Młoteczno brooch could be among the brooches from Gepid areas on the lower Danube. As an argument for the northern origin of this brooch, Cieśliński uses fragments of silver neck-ring with gold figurines of two embossed snakes from public building I in Gudme on Funen island from ca. mid-fifth century (Østergaard Sørensen 1994: 35 ff., fig. 14; Cieśliński 2010: 165). However, the publisher of the Gudme building underlines that several gold ornaments from it including gold linula pendant originate from the south-eastern Europe (Østergaard Sørensen 1994: fig. 11). The Młoteczno brooch is made in the Danubian style from phase D2 (Tejral 1988: ; Tejral 1997: 335f.; Tejral 2007: 65 72; Kazanski 1989; Kazanski 1996), probably its earlier stage, i. e. late fourth or early fifth century. This period is typical of semi-precious stone insets, knobs on head-plate, and framing with insets along the foot-plate sides. Such brooches basically concentrate in the middle Danube area, around Seven Castles of Transylvania (Siebenbürgen), and in the Crimea (Schmauder 2002: fig. 13; Fundliste 22). Bronze gilt brooch from Kalotė / Collaten in the area of Klaipeda belonged to the same horizon (Гавритухин 2000: 285, fig ), though it did not have polychrome insets and, according to survived drawings, had embossed decoration (Bezzenberger 1897: fig. 85; Кулаков 2009: 75, fig ). The design of 8-shaped wires decorating the knobs on the head-plate of the Młoteczno is extremely important because the same motif appears on two gold pendants from the same find. The similarity is so big that all three artefacts could be considered manufactured in the same workshop, as Peiser supposed long ago (Peiser 1919: 95). Similar pendants discovered in Cejkov and 8-shaped design on the brooch from Slovakian Ostrovány (Beninger 1931: 185 ff., pl. 9. 4; Prohászka 2006: fig. 33) from phase С2 suggest this ornamentation to appear earlier, though there also are later pendants of the kind, such as the find from the cemetery layer in Luboszyce (Domański 1979: fig. 24. k; Domański 1982: pl. 26. j) or another one from Solontsy in Western Ukraine (Котигорошко 1987: рис ; 8). Many gold and silver bucket-shaped pendants with granulation and filigree originate from the Przeworsk culture area, being Ines Beilke-Vogt s type N IIb, basically from phase C2 / C3 (Beilke-

261 A Gold Brooch from Młoteczno (Hammersorf), Braniewo Powiat, in North-Eastern Poland 261 Fig. 14. Årslev (Storgaard 1994) Vogt 1998: 56, 71, fig. 61, Fundliste 39; Bursche 1998: 145f.). The same 8-shaped ornament occurs on eyelets of two medallions of Valens from Szilágysomlyó (Bursche 1998: 146; Dembski 1999: Kat. 8, 9). Granulated gold pear-shaped pendant from the hoard of Grushevtsy in Bukovina (all but three artefacts disappeared) with vessels probably similar to those from Szilágysomlyó (Маркевич, Рикман 1973: рис. 1; Bursche 1998: 147). Moreover, this 8-shaped motif repeats on two silver gilt brooches from the hoard of Zamość (Kokowski 1995: kat , ryc. 63; Merowingerzeit 2007: Kat ; Fig. 4). Both these hoards, Szilágysomlyó and

262 262 Magdalena MĄCZYŃSKA Fig Lago di Varese; 2 Kerch (Germanen 1987; Ajbabin 1994) Zamość, date from the late fourth to the mid-fifth century, or to phase D2, and the latest elements of the Szilágysomlyó find to the mid-430s, i. e. phase D2/D3. 4 Therefore, gold pendants from Młoteczno should not be the earliest of the entire find, perhaps they are synchronous with the brooch, as it comes from the above-discussed ornamentation of both categories of artefacts, which nevertheless existed for long, from phase С2 to D2. Cieśliński also suggests that the pendants are later than it was considered earlier (Cieśliński 2010: 162). Let us return to the intriguing origin of the Młoteczno brooch and its place in the sequence of brooches of types Zakrzów and Untersiebenbrunn / Szi lágysomlyó (cf. above). Gold brooches А 184 with granulation and filigree from all the three burials in Wrocław-Zakrzów (Grempler 1888: pl. V. 9, 10; Grempler 1888a: pl. III. 1, 3; VII. 1 3; Quast 2009: fig , 3; 64. 1, 2) are a little bit earlier than the brooch with high receiver and high cord from Stráže (Kolník 1964, fig. 2). In phase С2, brooches with rich decoration of granulation and filigree appear in the Leuna Hassleben Gommern horizon, for example, the finds from Hassleben (Schulz 1933: pl. 4. 1a, 3a) or Ostrovány (Prohászka 2006: pl. 31, 33). Two particular find of such brooches, one with carnelian and glass insets, originate from Hungary (Hampel 1905 Bd. 2: 11 ff., 51; Hampel 1905 Bd. 3: pl. 12, 43. 8), dating back to the same period according to the style. Recent studies consider two gold brooches from Funen (Sanderumgård and Årslev) from phases С2 and С3 respectively (Andersson 1993: Kat b; 404. i; Jørgensen, Petersen 1998: 176, 178; fig ) to be local products (Andersson 1995: 221 ff.). They, as well as the brooch with carnelian inset from Årslev, accompanied with gold lion pendants, underline the acquaintance with Eastern Europe (Storgaard 1994; Quast 2011: 199 ff.). As for the filigree artefacts from Wrocław-Zakrzów, especially from grave II which particularly contained lunula pendant, buckle with carnelian inset, and cut glass beaker, Dieter Quast has interpreted them as a link to the Chernyakhov culture, supposing that the lady buried in that grave, who was wife to a local prince or king, probably was of the northern Black Sea origin (Quast 2011: 203 ff.). In the light of the mentioned finds of gold brooches from the late Roman period and in contrast to Stark (Stark 1999: 142) and Hilberg (Hilberg 2009: 161), I am not convinced in purely 4 See recent studies: Tejral 1997: 335 f.; Harhoiu 1998: 97; Bursche 1998: 51; Kiss 2001: 224 ff.

263 A Gold Brooch from Młoteczno (Hammersorf), Braniewo Powiat, in North-Eastern Poland 263 Fig Gelénes; 2 Szilágysomlyó (Kiss 1999) regional, Scandinavian usage of rich ornamentation with granulation and filigree. In phase D2 and earlier, especially in the period when the so-called Danubian style was popular, different styles mixed with a huge participation of Pontic elements. As it is indicated by a pair early silver gold-clad two-plate brooches with semicircular head-plate from Lago di Varese (Werner 1961: 26, Kat. 87, Taf. 19; Germanen 1987: Kat. I. 5. c), two gold brooches from Gelénes, two pairs of brooches from Szilágysomlyó (Kiss 1999: Kat ) with rich granulation, and brooches with filigree and garnets from two burials of in Kerch (Засецкая 1993: кат. 90, табл. 23; Ajbabin 1994: Kat. II. 1. k), this decorative pattern still existed in the Great Migration period, at least as long as the mid-fifth century. Big conical ornamented spring axle terminals and knob on the head-plate related to the Pomeranian brooches also have, however very distant, parallels among the sea-horse brooches from Szilágysomlyó. Nevertheless, general appearance and style of the Młoteczno brooch make it closer to the Danubian finds in contrast to Scandinavia and Pomerania, therefore I would better consider it an import from a centre of art located in the lower or middle Danube area. 5 Not far to the west of the main concentration of gold artefacts in Młoteczno, there was a partly destroyed Western Baltic cemetery, investigated in 1970s by Włodzimiera Ziemlińską-Odojowa (1991), which had existed in phases C3, D, and E (Cieśliński, Nowakowski 2005: 262 ff.; Cieśliński 2010: 167). However, gold artefacts from Młoteczno are of purely Germanic appearance, so the proximity to the Western Baltic cemetery suggests the coexistence of both tribes. Although we still not know for sure the exact place where the Młoteczno brooch was made, it was found east of the lower Vistula, within Germanic and not Western Baltic context from the final stage of the Wielbark culture or even after its disappearance, which was an evidence of the polychrome style distribution far to the north and of its influence on the artisans working in the southern Baltic Sea area. 5 Instead, the mentioned brooch from Kalotė / Collaten with stamped design could be of local origin.

264 264 Magdalena MĄCZYŃSKA Bibliography Åberg N. Ostpreussen in der Volkerwanderungszeit. Uppsala, Ajbabin A. I Goti in Crimea (secoli V VII) // I Goti. Milano, Andersson K. Romartida guldsmide i Norden. I. Katalog. Aun 17. Uppsala, Andersson K. I Romartida guldsmide i Norden. III. Ovriga smycken, teknisk analys och verkstadgrupper. Aun 21. (Uppsala 1995). Annibaldi G., Werner J. Ostgotische Grabfunde aus Acquasanta, Prov. Ascoli Piceno (Marche) // Germania Beilke-Vogt I. Frühgeschichtliche Miniaturobjekte mit Amulettcharakter zwischen Britischen Inseln und Schwarzem Meer // Universitätsforschungen zur Prähistorischen Archäologie. 51. Bonn, Beninger E. Der Wandalenfund von Czeke-Cejkov // Annalen des Naturhistorischen Museums in Wien Bezzenberger A. Katalog des Prussia-Museums im Nordflügel des Königlichen Schlosses zu Königsberg i. Pr. 2. Königsberg, Bierbrauer V. Das Frauengrab von Castelbolognese in der Romagna (Italien). Zur chronologischen, ethnischen und historischen Auswertbarkeit des ostgermanischen Fundstoffs des 5. Jahrhunderts in Südosteuropa und Italien // Jahrbuch des RGZM Bierbrauer V. A Luxury Brooch from the Second Szilágy-Somlyó Treasure? // From Attila to Charlemagne. Arts of the Early Medieval Period in The Metropolitan Museum of Art. New York, Bott H. Zur Datierung der Funde aus Hammersdorf (Ostpreussen) // Jahrbuch des RGZM Bursche A. Złote medaliony rzymskie w Barbaricum. Symbolika prestiżu i władzy społeczeństw barbarzyńskich u schyłku starożytności (Światowit. Supplement Series A: Antiquity 2). Warszawa, Cieśliński A. Die spätrömische Kaiserzeit und die frühe Völkerwanderungszeit an der Passarge / Pas łęka // The Turbulent Epoch. New materials from the Late Roman Period and the Migration Period / Monumenta Studia Gothica. 5. Lublin, Cieśliński A. Kulturelle Veränderungen und Besiedlungsabläufe im Gebiet der Wielbark-Kultur an Łyna, Pasłęka und oberer Drwęca // Berliner Beiträge zur Vor- und Frühgeschichte N. F. 17. Berlin, Cieśliński A., Nowakowski W. Die Passarge ein barbarischer Rubikon. Entwurf einer For schun gsproblematik // Wasserwege: Lebensadern Trennungslinien. Neumünster, Dembski G. Katalog, Nr // Barbarenschmuck und Römergold. Der Schatz von Szilagysomlyó. Milano; Wien, Domański G. Kultura luboszycka między Łabą a Odrą w II IV wieku. Wrocław, Domański G. Cmentarzysko w Luboszycach (II IV w.). Wrocław, Ebert M. Neuerwerbungen des Prussia-Museums. VI. Völkerwanderungszeit // Prussia Eggers H.-J., Stary P. E. Funde der Vorrömischen Eisenzeit, der Römischen Kaiserzeit und der Völker wanderungszeit in Pommern // Beiträge zur Ur- und Frühgeschichte Mecklenburg-Vorpommerns. 38. Lübstorf, Fettich N. Der zweite Schatz von Szilágysomlyó // Archaeologia Hungarica. 8. Budapest, Germanen, Hunnen und Awaren. Schatze der Volkerwanderungszeit. Nurnberg, Godłowski K. Okres wędrówek ludów na Pomorzu // Pomorania Antiqua Grempler W. Der Fund von Sackrau. Brandenburg an der Havel, Grempler W. Der II. und III. Fund von Sackrau. Berlin, 1888a. Hampel J. Alterthümer des frühen Mittelalters in Ungarn. Bd Braunschweig, Harhoiu R. Über die steinverzierten Fibeln mit halbkreisförmiger spiral- und rautenförmiger Fussplatte // Dacia N. S /39. Harhoiu R. Die frühe Völkerwanderungszeit in Rumänien // Archaeologia Romanica. 1. Bukarest, Hilberg V. Masurische Bügelfibeln. Studien zu den Fernbeziehungen der völkerwanderungszeitlichen Brandgräberfelder von Daumen und Kellaren // Daumen und Kellaren Tumiany i Kielary 2 / Schriften des Archäologischen Landesmuseums. 9. Neumünster 2009.

265 A Gold Brooch from Młoteczno (Hammersorf), Braniewo Powiat, in North-Eastern Poland 265 Hirschfeld G. Bruchstücke von zwei silbernen Geräten // Prussia Jørgensen L., Petersen P. V. Gold, Power and Belief. Danish gold treasures from Prehistory and the Middle Ages. Copenhagen, Kazanski M. La diffusion de la mode danubienne en Gaule (fin du IVe siècle début du VIe siècle): essai d interpretation historique // Antiquites Nationales Kazanski M. Les tombes princières de l horizon Untersiebenbrunn, le problème de l identification ethnique // Rencontres Internationales d Archeologie et d Histoire d Antibes Kiss A. Katalog, Nr // Barbarenschmuck und Römergold. Der Schatz von Szilagysomlyó. Milano; Wien, Kiss A. Die völkerwanderungszeitichen und frühmittelalterlichen Schatzfunde im Karpatenbecken // Roman Gold and the Development of the Early Germanic Kingdoms / Kungl. Vitterhets Historie och Antikvitets Akademien Konferenser Kokowski A. Schätze der Ostgoten. Stuttgart, Kolník T. Honosné spony mladšej doby rímskej vo svetle nálezov z juhozápadného Slovenska // Slovenska Archeologia La Baume W. Germanische Funde der Völkerwanderungszeit aus Nordostdeutschland // Blatter fur deutsche Vorgeschichte Levada M. To Europe via the Crimea: on possible migration routes of the northern people in the Great Migration Period // Inter Ambo Maria. Contacts between Scandinavia and the Crimea in the Roman Period / Cultural- Historical Reports. 10. Kristiansand; Simferopol, Mackeprang M. B. Aarslev-Fundet. Et rigt fynsk gravudstyr fra 4. aarh. e. Kr. // Fra Nationalmuseets Arbejdsmark Mangelsdorf G. Der Halsring von Peterfitz, Hinterpommern. Ein Beitrag zur zeitlichen und kulturellen Stellung der Goldringe mit verdickten und übergreifenden Enden des 6. Jahrhunderts in Skandinavien und südlich der Ostsee // Offa (2006/2007). 63/64. Merowingerzeit. Europa ohne Grenzen. Berlin, Munksgaard E. Spätantikes Silber // Frühmittelalterliche Studien Nordquist B. Der Kriegsbeuteopferplatz von Finnestorp in Schweden // Offa (2004/2005). 61/62. Østergaard Sørensen P. Gudmehallerne. Kogelit byggeri fra jernalderen // Nationalmusets Arbejdsmark Peiser F. E. Der Goldfund von Hammersdorf // Mannus Pilet Ch. Du Trésor d Airan á la tombe de la Princesse de Moult // Barbaren im Wandel. Beiträge zur Kultur und Identitätsumbildung in der Völkerwanderungszeit / Spisy AÚ AV ČR. 26. Brno, Prohászka P. Das vandalische Königsgrab von Osztrópataka (Ostrovany, SK) // Monumenta Germanorum Archeologia Hungariae. 3. Budapest, Quast D. Wanderer zwischen den Welten. Die germanischen Prunkgräber von Stráže und Zakrzów. Mainz, Quast D. The Links between the Crimea and Scandinavia: some jewellry from the third century AD princely graves in an international context // Inter Ambo Maria. Contacts between Scandinavia and the Crimea in the Roman Period / Cultural-Historical Reports. 10. Kristiansand; Simferopol, Salin E., France-Lanord A., Le trésor d Airan en Calvados // Monuments et mémoires Schmauder M. Oberschichtgräber und Verwahrfunde in Südosteuropa im 4. und 5. Jahrhundert // Archaeologia Romanica. 2. Bukarest, Schulz W. Das Fürstengrab von Hassleben. Berlin, Stark R. Die Fibeln // Barbarenschmuck und Römergold. Der Schatz von Szilagysomlyó. Milano; Wien, Storgaard B. The Årslev Grave and Connections between Funen and the Continent at the End of the Later Roman Age // The Archaeology of Gudme and Lundeborg / Arkæologiske Studier. 10. København, Svoboda B. Čechy v dobĕ stĕhovaní narodů // Monumenta Archaeologica. 13. Praha, Tejral J. Zur Chronologie der frühen Völkerwanderungszeit im mittleren Donauraum // Archaeologia Austriaca Tejral J. Neue Aspekte der frühvölkerwanderungszeitlichen Chronologie im Mitteldonauraum // Neue Beiträge zur Erforschung der Spätantike im mittleren Donauraum / Spisy AÚ AV ČR. 8. Brno, 1997.

266 266 Magdalena MĄCZYŃSKA Tejral J. Das Hunnenreich und die Identitätsfragen der barbarischen gentes im Mitteldonauraum aus der Sicht der Archäologie // Barbaren im Wandel. Beiträge zur Kultur und Identitätsumbildung in der Völkerwanderungszeit / Spisy AÚ AV ČR. 26. Brno, Werner J. Katalog der Sammlung Diergardt (Völkerwanderungszeitlicher Schmuck). 1. Die Fibeln. Berlin, Werner J. Zu einer elbgermanischen Fibel des 5. Jahrhunderts aus Gaukönigshöfen, Ldkr. Würzburg. Ein Beiträge zu den Fibeln vom Typ Wiesbaden und zu Punzornamentik // Bayerische Vorgeschichtsblätter Zeiss H. Ein Fibelfund des 5. Jahrhunderts von Schwellin, Kr. Köslin (Pommern) // Germania Ziemlińska-Odojowa W. Wstępne wyniki badań na wielokulturowym stanowisku w Młotecznie (gmina Braniewo) // Archeologia bałtyjska. Olsztyn, Айбабин А. И. Хронология могильников Крыма позднеримского и раннесредневекового времени // МАИЭТ I. Баранов В. И. Погребение гуннского времени у хутора Саги. К истории обнаружения и публикации находки 1887 г. // Славяно-русское ювелирное дело и его истоки. СПб., Гавритухин И. О. Финал традиции культур римского времени в Восточном Прикарпатье // Die spätrömische Kaiserzeit und die frühe Völkerwanderungszeit in Mittel- und Osteuropa. Łódź, Засецкая И. П. Материалы Боспорского некрополя второй половины IV первой половины V вв. н. э. // МАИЭТ III. Засецкая И. П. Культура кочевников южнорусских степей в гуннскую эпоху (конец IV V вв.). СПб., Котигорошко В. Г. Жертвенник III IV вв. н. э. у села Солонцы // СА Кропоткин В. В. Римские импортные изделия в восточной Европе II в. до н. э. V в. н. э. / САИ. Д М., Кулаков В. Находки из ареалов куршей и скальвов в описи копий Центрального Римско-Германского Музея (г. Майнц) // Achaeologia Lituana Маркевич В. И., Рикман Э. А. Клад III IV вв. н. э. из Буковины // СА Магдалена МОНЧИНЬСКАЯ Золотая фибула из Млотечно (Хаммерсдорф) Браневского повята в северо-восточной Польше Резюме В гг. в разных местах деревни Млотечно (бывш. Хаммерсдорф) было найдено не менее 17 золотых и серебряных предметов, из которых сегодня известны фрагменты двух серебряных сосудов восточноримского и галльского или италийского производства, медальон Констанция II ( гг.), золотая фибула, три золотые ведёрковидные подвески и две золотые гривны. Гривны относятся к VI в., а остальные находки могут быть датированы концом VI и первой половиной V в. Золотая фибула с гранатовыми и янтарными вставками вызвала оживлённую дискуссию, продолжающуюся уже несколько десятилетий. Основная проблема местного ли она происхождения или была изготовлена на среднем или нижнем Дунае. Стиль фибулы считается переходным между стилями «Закшув» и «Унтерзибенбрунн Силадьшомьё». О её местном происхождении писали К. Годловский, Р. Старк, А. Цеслиньский и Ф. Хильберг, также указывавшие на скандинавское влияние, проявившееся в использовании скани и зерни, а о дунайском Й. Вернер, Р. Хархою и М. Шмаудер. На основании стилистических признаков фи булы из Млотечно и собранных аналогий (фи бул типа «Нежин Силадьшомьё Ре гёль Ун терзибенбрунн Рабапордань» начала фазы D2), следует признать её продуктом дунайской мастерской, оказавшимся далеко на севере.

267 267 Boris MAGOMEDOV Correlation between North-Western and Local Elements in the Chernyakhov Culture According to the opinion of many modern scholars, the Chernyakhov culture is an archaeological reflection of the Gothic federation of different barbarian peoples (Казанский 2011: 22). However, the ratio of different cultural and ethnic elements is still disputable. The way to solve this problem is through the accumulation and study of the archaeological evidence: house-building, funeral rite, complex of artefacts, and through the search for similar evidence in geographically and chronologically close cultures, with necessary investigation of palaeoanthropological and historical data. 1 The colleagues interpret the Chernyakhov population as the Germans and Slavs, Sarmatians, Late Scythians and Thracians. The Germans are primarily the Goths, or the population of the Wielbark culture (Lubowidz phase) in the early Roman period. During the migration from the Baltic to the Black Sea, the Goths were joined by groups from other Germanic tribes. There were the Gepids, the most close relatives of the Goths (the population of late, Cecele phase of the Wielbark Culture), Vandals (the Przeworsk population), Herulians (who came from Jutland), Burgundians (the Luboszyce population), and possibly others. The Wielbark element is usually presented in the Chernyakhov sites with different groups of archaeological materials, often in the form common with other Germanic cultures in Central Europe. The most popular type of the Chernyakhov dwellings are pit-houses and timber frame huts, typical of the Wielbark settlements. The Wielbark feature of the Chernyakhov funeral rite is bi-ritualism (combining cremation and inhumation graves). The Wielbark elements also appear in the peculiarity of the cremation rite, as well as in predominance of inhumations in simple pits oriented to the north (80% of inhumations). The nobility graves in large pits with wooden chambers also have the parallels in the Germanic cultures. The Wielbark types of hand-made ceramics have been found on the most parts of the Chernyakhov territory, especially on the sites from the early period. They are few or absent in regions with small Chernyakhov population (the upper Dniester area), or where the Chernyakhov culture appeared later, when handmade vessels were generally supplanted by wheelmade ware (the Black Sea area, Wallachia, and Transylvania). The Chernyakhov style of pottery developed as a combination of the Roman provincial technology and forms and the Wielbark (sometimes Przeworsk) tradition of hand-made ware (fig. 1). Details of clothing, jewellery, household items, and the Wielbark types of weapons completely dominate in the Chernyakhov area and were borrowed by neighbouring Kiev and Alanic cultures. The early medieval Gothic women s costume passed through the Wielbark and Chernyakhvo stages of development (Tempelmann-Mączyńska 1989: 135). The characteristic Przeworsk feature different from the Wielbark and the Chernyakhov cultures is the presence of weapon graves according to cremation rite. These burials were found in eight Chernyakhov cemeteries in different regions as well. The wide distribution of three-handled vases (not typical of the early Wielbark sites) and a certain type of hand-made pots in the Chernyakhov culture is also connected with the Przeworsk (i. е. Vandalic) tradition. The presence of the Herulians is marked by the long houses, monster brooches, items with runic inscriptions, iron combs, and a certain type of pottery. According to Oleg Sharov, the eastern branch of the Burgundians participated in the shaping of the Chernyakhov culture. The Leuna Hassleben horizon of antiquities 1 The main points of this paper concerning the ethnic composition of the Chernyakhov culture have been stated in my monograph (Магомедов 2001).

268 268 Boris MAGOMEDOV Fig. 1. Types of the Przeworsk (1, 4) and Wielbark s (2, 3, 5, 6) hand-made vessels and their inheritance in the Chernyakhov wheel-made pottery (7 13). 1 Kamienczyk; 2 Brest-Trishin; 3 Rydzewo; 4 Garwolin; 5 Ditinichi; 6 Zagorzice; 7 Kamenka-Dneprovskaya; 8 Kosanov; 9 Chernyakhov-Romashki; 10 Danceni; 11 Voyskvoye; 12 Sosnova; 13 Petrikivtsy and several types of vessels appeared in connection with them (Шаров 1991). The Chernyakhov population could also include two groups of Iranian origin: the Sarmatians, with their leading tribe of Alans, and the Late Scythians. The most reliable archaeological features to highlight the Sarmatian-Alanic element in the Chernyakhov culture are the types of graves: undercut constructions and catacombs (in most regions they are about 1 % of the burials, though in the Black Sea area 20 %), and also inhumations with weapons (in seven cemeteries) and artificially deformed skulls (in four cemeteries). In the Black Sea and the lower Danube areas, there are pots of Sarmatian types. Metal mirrors from the Hunnic period (five finds) are also of the Sarmatian origin. The Sarmatian influence can be seen in the Chernyakhov women s costume as beaded ornaments discovered on legs in some cases. In the northern Black Sea and the lower Danube areas,

269 Fig. 2. Distribution of skull morphotypes in the Chernyakhov cemeteries. Regions and cemeteries. I. Western Ukraine: 1 Chistilov; 2 Bila; 3 Cherneliv-Ruskiy; 4 Romanovo Selo; 5 Petrikivtsy. II. Upper Dniester. III. Central Ukraine: 6 Velikaya Bugayovka; 7 Derev yana; 8 Chernyakhov; 9 Romashki; 10 Teleshovka; 11 Kurniki; 12 Kosanov; 13 Zhuravka; 14 Zavadovka; 15 Cherkassy-Centre; 16 Maslovo; IV. Eastern Ukraine: 17 Uspenka; 18 Sumy-Sad; 19 Lokhvitsa; 20 Boromlya; 21 Kantemirovka. V. Lower Dnieper. VI. Northern Black Sea area: 22 Nagornoye; 23 Kholmskoye; 24 Ranzhevoye; 25 Koblevo; 26 Viktorovka; 27 Kaborga; 28 Kamenka-Anchekrak; 29 Nikolayevka; 30 Gavrilovka; VII. Moldova: 31 Romankovtsy; 32 Mălăieşti; 33 Budeşti; 34 Bălţata; 35 Erbiceni; 36 Letcani. VIII. Wallachia: 37 Independenţă. IX. Transylvania: 38 Sântana de Mureş Conventional signs: 1 morphotype prevails; 2 morphotype is presented; 3 types of the Gothic circle; 4 Late Scythian type; 5 Sarmatian type; 6 Mediterranean type; 7 Slavonic type; 8 large cemeteries with unspecified morphotypes; 9 Greco-Roman cities and towns; 10 borders between the areas of archaeological cultures; 11 borders between the areas in the Chernyakhov culture; 12 Roman border in the fourth century

270 270 Boris MAGOMEDOV there are traces of the settled Sarmatian population among the Chernyakhov people, though on the rest of the Chernyakhov area such features appear in individual graves. The Late Scythians in the Chernyakhov community are anthropologically close to the population of the sites of lower Dnieper and the Crimea in the previous period. They had typical stone houses of a specific type (in the Black Sea area), funeral rite with western orientation of the dead, ledged graves, and bottom of graves plastered with a layer of green clay, and Late Scythian types of pots among the hand-made ware. Slavonic element in the Chernyakhov culture appeared in two groups of sites, western and eastern. The western group consists of settlements in the upper and middle Dniester, with no cemetery discovered near these settlements. The ceramic complex contains % of hand-made ware, with many types meeting with analogies in the early mediaeval Prague culture. Although pit-houses with stone stoves discovered in some of the sites have been considered an important Slavic feature (Славяне 1990: 127), an authoritative researcher Igor Gavritukhin has recently posed (but not published) arguments that they are from a later, early mediaeval period. In the Late Roman period, the Slavonic population in the east of the Ukraine and in the adjacent regions of the Russian Federation is presented by the antiquities of the Kiev culture. Its elements have been detected mainly in eight Chernyakhov settlements, basically in the middle Dnieper area, usually as single finds of ceramic vessels. The south-western part of the Chernyakhov area was previously occupied by the Thracian circle of the Carpians and Dacians. The finds of specific hand-made ceramics of Thracian types are known in some Chernyakhov settlements and cemeteries in Romania. Let us supplement the archaeological observation with palaeoanthropological data (Магомедов, Рудич 2008). Most of the genetic connections of the Chernyakhov population are directed to the north-west, i. e. to Central and Northern Europe (fig. 2). Such skulls include morphotypes similar to some Germanic populations, the Celts and the Balts. The ethnic core of Goths migrated to the southern Baltic Sea area from Scandinavia, joined by representatives of different tribes. Therefore, the Chernyakhov Goths have a complex anthropological structure, including the North and East Germanic elements, as well as the descendants of the assimilated Celts and Balts. In the Chernyakhov context, these anthropological types could be considered a Gothic circle. Skulls with the features considered typical for different territorial and chronological groups of the Sarmatians make rarely appear in the Chernyakhov cemeteries. In various areas, these skulls are 5 to 12 %. The Late Scythian morphotype exists in the most Chernyakhov sites in the Black Sea area, dominates in some Central and Eastern Ukrainian cemeteries in, and it is less common in other regions. An important result of new research is the disclosure of real anthropological traces of Slavic ethnos in the Chernyakhov culture (Рудич 2007). Such features have been documented in small series of female skulls from the cemeteries of Cherneliv- Rusky (Western Ukraine) and Nagornoye 2 (near the Danube estuary). This type is similar to the women s series of the Maslomęcz group of the Wielbark culture (Maslomęcz, Gródek). Mediterranean anthropological type has been revealed in the Black Sea cemeteries in Moldavia, Muntenia, Western and Central Ukraine. The criteria allowing the exact determination of the Greeks and Thracians, as well as the Romans, are not developed so far. Therefore, anthropologists discuss the ethnic roots of the Mediterranean group of the Chernyakhov population against the background of archaeological data. Conclusion The homogeneity of the majority of the Chernyakhov sites can be explained by the homogeneity of its population, represented mainly by the Goths, formerly the people of the Wielbark culture. Some local peoples formed quite large ethnic arrays on the territory of the Chernyakhov culture. This is the case of the Western Slavonic group in the upper and, partly, middle Dnieper area, and, to a lesser extent, of the Late Scythians if the northwestern Black Sea area. Cultural elements of the Late Scythian group, Slavs of the Kiev culture, and Carpo-Dacians appear, as a rule, on the sites created by the main population bulk.

271 Correlation between North-Western and Local Elements in the Chernyakhov Culture 271 Bibliography Tempelmann-Mączyńska M. Das Frauentrachtzubehör des mittel- und osteuropäischen Barbaricums in der römischen Kaiserzeit. Kraków, Казанский М. М. Радагайс и конец черняховской культуры // Oium. 1. Київ, Магомедов Б. В. Черняховская культура. Проблема этноса / Monumenta Studia Gothica. I. Lublin, Магомедов Б. В., Рудич Т. O. Некоторые антропологические аспекты черняховской культуры // The Turbulent Epoch. New Materials from the Late Roman Period and the Migration Period. I. Lublin, Рудич Т. О. Роль населення черняхівської культури у формуванні антропологічного складу слов янської людності України другої половини І початку ІІ тис. н. е. // Археологія Славяне Юго-Восточной Европы в предгосударственный период. Киев, Шаров О. В. Бургунды в Подунавье и Северном Причерноморье // Древнейшие общности земледельцев и скотоводов Северного Причерноморья. (Материалы конференции). Кишинёв; Киев, Борис МАГОМЕДОВ Соотношение северо-западных и местных элементов в черняховской культуре Резюме Однородность большинства черняховских памятников можно объяснить гомогенностью оставившего их населения, представленного в основном готами в прошлом носителями вельбарской культуры. При переселениях готов из Прибалтики в Причерноморье в их состав вливались группы людей из других германских племён. Некоторые местные народы составляли на черняховской территории достаточно крупные этнические массивы. Это западная группировка славян на верхнем и среднем Днестре, а также, в меньшей степени, поздние скифы Северо-Западного Причерноморья. Культурные элементы части позднескифского населения, славян киевской культуры и карпо-даков встречаются, как правило, на памятниках основного населения. Археологические наблюдения подтверждаются данными антропологии.

272 272 Anna MASTYKOVA, Mariya DOBROVOL SKAYA Grave Goods in the Cemeteries from the Late Roman and Early Mediaeval Periods on the Sambian Peninsula and Anthropological Accounts The attribution of the sex and age of the Sambian-Natangian 1 culture burials in the Late Roman period, Great Migration, and Early Middle Ages is topical for the archaeology of the Western Balts. The rite of cremation dominates in the cemeteries in Sambia and Natangia from those periods, and no attribution of bone remains has been done so far. Therefore, almost all the age-sex attributions are based on grave goods, which, according to generally accepted experience of the Early Mediaeval Baltic, Slavonic, and Germanic archaeology, are determined as man s if there are weapons, horse trappings, the so-called warriors belt sets, or woman s with ornaments and details of costume, for example beads, pairs of brooches, and artefacts related to home production, such as spindle whorls. These purely hypothetical conclusions form a usual background for the deductions about the chronological evolution of local costume or social stratification of ancient Balts. However, it is worth notice that only a small part of archaeological monuments from eastern Prussia has been published and just a few cemeteries have become subject of palaeoanthropological research so far. Therefore, topical are the results obtained by palaeoanthropological researches (Добровольская 2010) of the materials from the early mediaeval cemetery of Mitino (Gur yevsk district in the Kaliningrad oblast administrative area) studied by Konstantin Skvortsov (Скворцов 2010a; 2010b). Methodological approaches The research of cremation graves uses both methods developed by forensic medical examination and archaeological approaches (for example: Stewart 1979; Holck 1996). The general observation of the materials has suggested that the colour of the fragments is stable. Whitish grey colour and size of the fragments (0.5 to 5 6 cm in diameter) allows us to suppose that the cremations sustained almost the same temperature and period of burning. The probable temperature was at least º C (Walker, Miller, Richman 2008), though the burning time was not shorter than one hour. A few bone accumulations of small weight (less then 10 g) and with very fine fragments do not allow the age and gender attribution. Although some cremations weighs low because the individual was of the child s or baby s age of the individuals, for such an attribution there must be cranial bone fragments with sutures, parts of cranial vault with under-developed diploe, orbital parts of frontal bone, epiphysis areas of long tubular bones, etc. When we are dealing with accumulated small fragments without child s features, we can also infer that these fragments developed because of long burning of the body on funeral fire; anyway, such a suggestion should be checked every next time. The accumulations almost not contained nonhuman materials but, in a few cases, tiny pieces of coal (less than 0.5 cm in big diameter) and possibly miniature crumbles of chalk. These features suggest that bone fragments were brought to graves from burning sites and were carefully cleaned but perhaps not washed. The age and sex are identified by anthropologically investigable fragments: sex by such features as the relief development of front, temple and back 1 Polish archaeology calls these sites the Dollkeim-Kovrovo culture, though Russian archaeology prefers the name of the Sambian-Natangian culture (group), as it has been suggested by Jerzy Okulicz.

273 Grave Goods in the Cemeteries from the Late Roman and Early Mediaeval Periods on the Sambian Peninsula and Anthropological Accounts bones of skull, the forms of mental protuberance, low mandibular angle and upper outside line of orbit, the parietal bone solidity, and the size of epiphysis of tubular bones. Obviously, the reliability of age-sex attribution depends on the preservation of fragments that remained. The probability that one or few anatomically determinable fragments appear in a large accumulation of cremated remains is much higher than of meeting them among small isolated fragments. Therefore, almost every sex determination of persons belongs to accumulations of bones weighing more than 100 g, though all the attributions of sex of less-than-10-g accumulations are poorly reliable. Determination of age within wide intervals is possible with accumulation of small weight too. Such attributions evaluate the general solidness of various parts of skeleton, cases of osteoporosis, traces of unfinished synostosis, condition of articular surfaces of upper and lower extremities, upper and lower surfaces of vertebral bodies and cranial sutures, formation of tooth crown, root closure for the teeth, and the tooth crown wear. Mitino cemetery (fig. 1. 1) The excavations of the cemetery uncovered 408 sites with 275 (including 259 graves) from the late fifth to fourteenth century. Cremation graves in Mitino cemetery date from the late fifth to the early eighth century; they form a rather standard type of remains of cremations made somewhere outside and buried into pits in soil often covered with stonework. There were few cases when the graves were accompanied with burials of horses placed either near human grave or, more often, below it, in a grave pit (Скворцов 2010a: 16 31). The anthropological analysis has attributed bones from 73 graves from the late fifth to the early eighth century (Добровольская 2010: ) as, with high or small degree of probability, 30 men s, 21 women s (sometimes with children), and 22 children s graves (tab. 1) The comparison of palaeoanthropological data with the composition of grave goods allows us to determine the artefacts and funeral rite features typical of men s and women s graves. The most important in this case is the fact that Mitino cemetery is a rather ordinary site, that is to say, it reflects the common culture of the main bulk of the Sambian-Natangian population and therefore could be a kind of reference point for further investigations of the antiquities from the late fifth to the early eighth century in the Sambian peninsula. Although the number of graves of men, women, and children (the latter with no gender separation) was almost equal, so the comparison was correct, the total number of cremation under anthropological research makes our conclusions preliminary. It should be noted in regard to the composition of grave goods that burials of children contained the same artefacts and followed the same funeral rite as adults burials. The anthropological analysis included almost the same number of graves not accompanied with goods but ceramic ware (it appears in the absolute majority of graves): 6 (5) 3 men s graves, i. e. ca. 20 (17) % of total number of graves determined as men s: m2.13 4, m5.55 (?), m10.53, m13.39, m13.56, m44.377; 4 (4) women s graves, i. e. ca. 19 %: f6.22, f9.309в, f9.325, f17.341в 5 ; 7 (4) children s graves, i. e. 32 (18) %: ch3.140?, ch5.82, ch5.93, ch7.75?, ch9.306, ch13.48?, ch A great part of categories of grave goods is common for all three groups of burials (of men, women, and children) (fig. 2. А), with the number of graves with these goods sometimes varying in dependence of sex/age of the deceased, so one can determine some trends for the composition of grave goods. Single crossbow fibula (fig. 2. 3) was discovered in: 8 (8) men s graves, i. e. ca. 27 %: m9.274, m13.29, m17.217, m17.226, m17.321, m20.319, m20.329, m20.336; 2 The table is compiled according to the data amended after the cemetery of Mitino had been published in 2010 (Добровольская 2010: ). 3 The first number indicates the total quantity (or percentage) of the burials attributed to the given sex/age, brackets contain the quantity of graves with big mass of remains, i. e. with high degree of reliability. 4 Henceforth ch refers to child, f to woman, m to man,? to tentative anthropological attribution; then follow numbers of the trench and the grave. 5 The catalogue of graves in the cemetery of Mitino does not include the description of grave в, though there is its detailed plan indicating spots where goods, only pottery fragments, were uncovered (see: Скворцов 2010b: 683, табл. DLXXVII).

274 274 Anna MASTYKOVA, Mariya DOBROVOL SKAYA Fig. 1. The cemeteries in the Sambian peninsula mentioned in this paper. 1 Mitino, 2 Schlakalken-5 4 (3) women s graves, i. e. ca. 19 (14) %; f5.65?, f5.84, f22.119, fch ; 4 (3) children s graves, i. e. ca. 18 (14) %; ch17.304, ch5.76?, ch13.86, ch As we can see, single fibula appeared in comparable percentage of men s and women s graves, yet a bit more often in men s graves. A pair of crossbow fibulae (fig. 2. 4) was discovered in: 6 (6) men s graves, i. e. ca. 20 %; m17.262, m20.259, m20.332, m22.365, m20.318, m20.301; 4 (3) women s graves, i. e. ca. 19 (14) %. f13.57?, f17.207, f20.273, f Therefore, pairs of fibulae occurred only in graves of adults, as often in women s graves as in man s. Single beads (fig. 2. 1) were discovered in: 7 (7) men s graves, i. e. ca. 23%; m13.29 amb., 7 m9.274 amb., m amb., m amb., m amb., m amb., m amb.; 5 (4) women s graves, i. e. ca. 24 (19) %; f5.65? 2 amb., f amb., f13.61 amb., f17.341с amb., f amb.; 4 (2) children s graves, i. e. ca. 18 (9) %. ch5.89? amb., р5.348 amb., р amb. and gl., ch ? amb. Such a traditional artefact of woman s attire as beads was in graves of every kind (as scattered items) in almost equal number, yet beads were rare in graves of children. Buckles (fig. 2. 2) were discovered in: 8 (8) men s graves, i. e. ca. 27 %; m13.29, m17.321, m17.262, m20.259, m20.336, m sp., m20.301, m (attribution of this artefact as a buckle is disputable); 7 (7) women s graves, i. e. ca. 33 %; fch5.167 accumulation of calcined bones no. 1, 8 f22.119, 6 Henceforth fch refers to grave of a woman with a child. 7 Amb. refers to amber and gl. to glass beads. 8 Grave possibly contained remains of a woman and a child because the excavation uncovered two sets of calcined bones (Скворцов 2010b: 28), but the ratio of cremated remains in these accumulation has not been specified, so we consider grave to be that of a woman with a child.

275 Grave Goods in the Cemeteries from the Late Roman and Early Mediaeval Periods on the Sambian Peninsula and Anthropological Accounts f13.61, f17.341с, f20.273, f17.253, f20.303; 1 (1) child s grave, i. e. ca. 4.5 %. ch9.309а. In this case, there is an obvious trend for buckles to occur in graves of adults, both men and women. Knives (fig. 2. 5) were discovered in: 17 (17) men s graves, i. e. ca. 57 %; m5.35, m9.274, m10.45, m13.29, m13.52, m17.217, m17.226, m20.301, m two specimens, m20.319, m20.329, m20.332, m20.336, m22.365, m32.121, m44.390, m22.366; 12 (11) women s graves, i. e. ca. 57 (52) %; f5.65?, f5.84, fch5.167 accumulation no. 1, f22.119, f13.61, f17.207, f17.349, f20.273, f20.300, f17.253, f20.303, f5.155; 4 (3) children s graves, i. e. ca. 18 (14) %: ch1.288, ch5.76?, ch20.265, ch5.47. Knife in grave looks like a typical feature of adult burial, though does not form a doubtless indicator. There were some artefacts related mainly with graves of men (fig. 2. B). Spurs (fig. 2. 8) were discovered only in 2 (2) men s graves, i. e. ca. 7 %. m13.29, m Strap-ends (fig. 2. 6) occurred only in 2 (2) men s graves, i. e. ca. 7 %. m20.259, m Bracelet-like rings (neck-rings?) 9 (fig. 2. 7) were discovered in: 4 (4) men s graves, i. e. ca. 13 %: m13.29, m20.319, m20.329, m20.336; 1 (1) woman s grave, i. e. ca. 5 %; f20.300; 1 (1) child s grave, i. e. ca. 4.5 %. ch Horse s grave, often with elements of tack and harness (fig. 2. 9), accompanied the main burial in: 9 (9) men s graves, i. e. ca. 30 %; m2.158, m5.35, m13.29, m17.216, m17.217, m А 10, m20.259, m20.319, m44.390; 3 (3) women s graves, i. e. ca. 14 %; f22.119, f17.253, f20.303; 3 (2) children s graves, i. e. ca. 14 (9) %. ch1.288, ch5.76?, ch Some elements of grave goods probably indicate graves of women (fig. 2. C). 275 Pins (fig ) are discovered in: 3 (3) women s graves, i. e. ca. 14 %: f20.303, f17.341с, f ; 1 (1) child s grave, i. e. ca. 4.5 %: ch As we can see, so far pins were discovered in graves of women and a child. Three fibulae (fig ) were discovered in: 2 (2) women s graves, i. e. ca. 9.5 %; f20.300, fch5.167 accumulation no ; 1 (1) child s grave, i. e. ca. 4.5 %. ch Spindle whorls (fig ) were discove - red in: 4 (3) men s graves, i. e. ca. 13 (10) %; m5.67?, m17.217, m22.365, m20.318; 4 (4) women s graves, i. e. ca. 19 %; f13.61, f17.207, f17.349, f20.273; 2 (2) children s graves, i. e. ca. 9 %. ch44.388, ch In other words, women s graves more often contain spindle whorls, though spindle whorl does not form it an indisputable indicator. Pieces of untreated amber appear in one child s grave so far: р35.166?. Therefore, we can affirm that men s burials in Mitino cemetery are notable for the presence of spurs, strap-ends, and bracelets (neck-rings?) among grave goods, as well as horse s burials to accompany the main grave. Moreover, one fibula is also more typical of man s grave. The features of women s burials are not so pronounced. Nevertheless, we can suppose that three crossbow fibulae, pins, and ceramic spindle whorls among grave goods most likely indicate burial of a woman. Certainly, the cemetery of Mitino is so far almost the only site with a large series of cremations studied by palaeoanthropologist. Ho we ver, new investigation of the cemetery of Schlakalen-5 from the Late Roman period (in Zelenograd district of the Kaliningrad oblast, bet ween the modern settlements of Romanovo and Zaostrov ye, or Schlakalken, Kreis [district] Fischausen before 1945) confirms the results obtained by the study of Mitino cemetery (Мас ты ко ва 2011: ). 9 According to the data in possession (preservation, dimensions, location in grave), bracelet-like rings could not be undoubtedly interpreted as bracelets or neck-rings, so we combined them into a single category for now. 10 Grave А belonged to a horse and was located west of man s grave (Скворцов 2010b: 74 75, 639, табл. DXXXII). 11 The publication suggests that this artifact was a needle (Скворцов 2010b: 93, табл. DXLII. 2, 2a). However, the drawing suggests that it is rather a head of staff-shaped pin. 12 As for the attribution of grave fch5.167 see note 7 above.

276 276 Anna MASTYKOVA, Mariya DOBROVOL SKAYA Fig. 2. Features of grave goods and funeral rite typical of the cemetery of Mitino. А features common for graves of men and women; В features typical of graves of men; С features typical of graves of women. 1 beads; 2 buckles; 3 single fibula; 4 two fibulae; 5 knives; 6 strap-ends; 7 bracelet-like rings (neck-rings?); 8 spurs; 9 horse tack and harness, and burials of horses; 10 pins; 11 spindle whorls; 12 three fibulae (drawings based on Скворцов 2010a; Скворцов 2010b)

277 Grave Goods in the Cemeteries from the Late Roman and Early Mediaeval Periods on the Sambian Peninsula and Anthropological Accounts Schlakalken-5 cemetery (fig. 1. 2) The research of the cemetery of Schlakalken-5 discovered six flat graves of cremation rite (four in and two without urns). 13 Grave goods such as hand-made vessels, Roman coins, iron fibula with returned foot of Oscar Almgren s type 162 (Almgren 1897: Taf. VII. 1), iron spear-head, and ceramic spindle-whorls date this cemetery back to the Late Roman period, i. e. phases С1 С3 (ca. AD 160/ /370) according to the European Barbaricum timeline (see: Shchukin, Kazanski, Sharov 2006: v), generally corresponding to Wojciech Nowakowski s phases 2 4 of the chronology of the Sambian peninsula antiquities (Nowakowski 1996: 50 53). Anthropological research has analysed cremated materials from all the six graves in this cemetery. Grave 1. According to the sum of anthropological features, of a woman (younger than forty). Grave 3. The remains probably belonged to an adult male person. Grave 6. Sex and age were not determined. Unfortunately, all the listed graves unfortunately contained only insignificant and therefore nonattributable fragments of metal ware, though the materials from three other graves in the cemetery of Schlakalken-5 provided interesting results. Grave 2. The excavation uncovered brooch of Almgren s type 162 and bones of a man 30 to 50 year old. Bone materials for the analysis were taken separately from the infill of the urn, burial pit, and the accumulation outside the burial and urn. The results of the anthropological research evidence that all the three accumulations of bones in grave 2 belong to the same adult male person. Let us also mention iron spear discovered not far from the grave, which had possibly been driven off the grave by ploughing. The materials of this grave as well as the abovecited account of the cemetery of Mitino make us to suppose that the cremations in the Late Roman and Early Mediaeval periods with one fibula/brooch in the Sambian-Natangian culture area could belong to women or men, with this feature more typical of men s graves. Grave 4. It contained spindle whorl, so cremated remains could be of a woman s or child s skeleton. 277 Grave 5. The excavations discovered broken spindle whorl, so it was a burial of adult woman 40 to 50 year old. It is worth mentioning that in the cemetery of Mitino graves with spindle whorls most often belonged to women (see above). *** Although the selection of anthropological data from Schlakalken-5 cemetery is certainly very small, it supplies evidence that the conclusion regarding the relation between the set of grave goods and anthropological data made for Mitino cemetery could work for other monuments also. Besides, these materials show that the determination of man s and woman s indicators among the grave goods in the Great Migration and Early Mediaeval periods are probably correct for the Late Roman period as well. In this regard, it is interesting to compare the studies of Sambian cemeteries with palaeoanthropological accounts of the Western Baltic cemeteries in the neighbouring areas, particularly Netta in the Augustów group of the Sudovian culture from the Late Roman and the Great Migration periods, that is from the second half or the end of the second century BC (phase B2/C1 C1 in the European Barbaricum timeline) to the second half of the fifth or early sixth century (phase D/E) (Bitner- Wróblewska 2007), with anthropologically attributed cremations of at least 51 men, 56 women, and 49 children with grave goods of interest (Wiercińska 2007). It comes out that the distribution of demonstrative categories of artefacts in graves of men, women, and children is similar but not identical with what has been recorded in Sambian graves. This way, spurs and accompanying burials of horses have been determined only in man s grave 30, though strap-ends, typical of men s graves in Sambia, in the cemetery of Netta appear both in burials of men (grave 22, 175?), woman (grave 42В), and children (grave 24, 78). As for the typical women s grave goods, the picture is similar: pins, common in women s cremations in Sambia, occurred in two tentatively men s cremations (graves 142? и 175?) and graves of women (20 and 105?), twin burials of a woman and a child (61, 89, 100), and graves of 13 This site was discovered by the finds of Roman coins in the ploughed layer. We are grateful to Edvin Zal tsman (Kaliningrad State University) for this information.

278 278 Anna MASTYKOVA, Mariya DOBROVOL SKAYA children (93А, 102, 143, and possibly 94В). Spindle whorls, being a typical element of female grave goods in Sambia, are absent in graves of men in the cemetery of Netta, but occurred in female cremations (graves 59, 115, 25?, 107?), twin graves of a woman with a child (61, 73В, 91A D), and rarely in burials of a single child (23А, 78). Finally, knives, buckles, isolated beads, and single fibulae in the cemetery of Netta appear in burials of men, women, and children. Two fiulae are discovered in two twin graves, of a man with a child (22) and of a woman with a child (grave 138А), and in single burial of a woman (57), but are absent in single child s burials. We hope that further accumulation of palaeoanthropological data both in the Sambian peninsula and in its vicinity would help us to clarify or, possibly, to reconsider traditional attribution of the grave goods among the Western Balts. Appendix Table 1 Sex and age of the persons buried in cremation graves in the cemetery of Mitino Nos. of trenches men weight and burials 2 13 older than 35 less than 100 (eye-socket) older than 45 about younger than 40 less than adult less than years old more than older than 35 less than years old less than years old about years old less than years old about years old more than years old more than years old more than years old less than older than 40 more than years old more than years old more than years old more than years old more than years old more than years old more than older than 30 more than older than 40 more than older than 35 more than older than 30 more than years old adult less than adult about years old about adult less than 100

279 Table 2 Nos. of trenches and burials women weight years old more than younger than 40 less than years old less than в years old about years old, and a child about older than 45 more than years old more than years old about с years old about years old more than в years old more than years old more than years old more than years old more than years old more than years old more than adult less than years old less than years old about adult less than years old less than 100 Table 3 Nos. of trenches and burials children weight child less than child more than child less than child, 3 10 years old more than child less than child about а child less than child-adolescent about child less than child, less than 5 years old less than child, less than 5 years old less than child about child about child about child less than child less than child? less than child? less than child? less than child? about child? about child? less than 10

280 280 Anna MASTYKOVA, Mariya DOBROVOL SKAYA Bibliography Almgren O. Studien über Nordeuropäische Fibelformen der ersten nachchristlichren Jahrhunderte. Stockholm, Bitner-Wróblewska A. Netta. A Balt Cemetery in Northeastern Poland. Warszawa, Holck P. Cremated Bones. Oslo, Nowakowski W. Das Samland in der römischen Kaiserzeit und seine Verbindungen mit dem römischen Reich und der barbarischen Welt. Marburg; Warszawa, Shchukin M., Kazanski M., Sharov O. Des Goths aux Huns. Le Nord de la mer Noire au Bas-Empire et à l époque des Grandes Migrations / BAR IS Oxford, Stewart T. D. Essentials of Forensic Anthropology. Springfield Walker P. L., Miller K. W. P., Richman R. Time, temperature and oxygen availability: An experimental study of the effects of environmental conditions on the color and organic content of cremated bones // The Analysis of Burned Human Remains. San Diego, Wiercińska A. Human bone remains from the cemetery at Netta, site I osteological analysis // Bitner-Wróblewska A. Netta. A Balt Cemetery in Northeastern Poland. Warszawa, Добровольская М. В. Результаты антропологического анализа материалов, происходящих из погребальных комплексов могильника Митино // Скворцов К. Н. Могильник Митино V XIV вв. (Калининградская область). По результатам исследований 2008 г. 1. М., Мастыкова А. В. Охранные раскопки на могильниках позднеримского времени Шлакалькен-5 и Шлакалькен-6 на Самбийской полуострове // Труды III (XIX) Всероссийского археологического съезда. Великий Новгород Старая Русса. I. СПб.; М.; Великий Новгород, Скворцов К. Н. Могильник Митино V XIV вв. (Калининградская область). По результатам исследований 2008 г. 1. М., 2010a. Скворцов К. Н. Могильник Митино V XIV вв. (Калининградская область). По результатам исследований 2008 г. 2. М., 2010b. Анна МАСТЫКОВА, Мария ДОБРОВОЛЬСКАЯ Погребальный инвентарь могильников Самбийского полуострова позднеримского и раннесредневекового времени и данные антропологии Резюме В археологии западных балтов достаточно остро стоит вопрос определения половозрастной атрибуции погребений самбийско-натангийской культуры римского времени и начала средневековья. Здесь доминируют захоронения по обряду кремации, определения костных останков из них не проводились. Поэтому практически все половозрастные атрибуции погребений базируются на анализе сопровождающего инвентаря. Последний определяется как «мужской» (например, оружие, предметы конского снаряжения, так называемые «воинские» поясные гарнитуры) или «женский» (украшения и элементы костюма, предметы, связанные с домашним производством), исходя из общеевропейского опыта «варварской» археологии. На базе таких сугубо гипотетических заключений зачастую делаются выводы о хронологической эволюции местного костюма или социальной стратификации древних балтов. Поэтому представляются актуальными результаты, полученные при палеоантропологических исследованиях, проведенных на материалах некрополей Митино и Шлакалькен-5 на Самбийском полуострове (рис. 1). Сопоставление палеоантропологических сведений и состава погребального

281 Grave Goods in the Cemeteries from the Late Roman and Early Mediaeval Periods on the Sambian Peninsula and Anthropological Accounts 281 инвентаря позволило определить, какие предметы или черты погребального обряда присущи мужским захоронениям, а какие женским и детским. Так, на могильнике Митино (Гурьевский район Калининградской области) были антропологически определены костные останки из 73 погребений конца V начала VIII в. На могильнике Шлакалькен-5 (Зеленоградский район Калиниградской области), датированного в пределах второй половины II второй половины IV в., антропологически исследовано шесть погребений. Выборка антропологических данных могильника Шлакалькен-5, безусловно, очень мала. Тем не менее, материалы некрополя Шлакалькен-5 свидетельствуют, что выводы о соотношении набора погребального инвентаря и антропологических показателей, сделанные для могильника Митино, могут быть распространены и на другие памятники. Кроме того, эти материалы демонстрируют, что определения «мужских» и «женских» индикаторов погребального инвентаря, выявленные для эпохи Великого переселения народов и начала средневековья, видимо, актуальны и для позднеримского времени (рис. 2). С целью выяснения степени взаимного сходства или различия были сопоставлены результаты, полученные на могильниках Самбии с палеоантропологическими данными западнобалтского могильника Нетта (августовская группа судовской культуры) позднеримского времени и эпохи Великого переселения народов. Оказалось, что распределение показательных категорий предметов в мужских, женских и детских могилах здесь сравнимо, хотя и не полностью идентично тому, что мы наблюдаем на самбийских могильниках. Дальнейшее накопление данных палеоантропологических исследований как на Самбийском полуострове, так и на соседних территориях позволит уточнить, а, может быть, во многом и пересмотреть традиционные атрибуции погребального инвентаря западных балтов.

282 282 Andrey OBLOMSKIY The Upper Don and the Northern Black Sea Areas ca. 500 AD: aspects of the trade route Since the main topic of the discussion at the conference Inter Ambo Maria is the contacts between the peoples of the Barbaricum between the Baltic and the Black Seas, the search for trade routes from the northern Black Sea area to the inland becomes topical. These routes certainly existed, so the aim of the given paper is to discuss archaeological materials capable of uncovering the way from the Azov Sea, upstream of the Don, as far as the Oka in the Hunnic period (late fourth and fifth centuries). Which are these materials? One would think of the finds of imported ware, of late Greco-Roman origin in our case, as the main feature of such a trade route. However, the overall picture and the dynamics of the distribution of Greco-Roman imports in the Roman and Hunnic periods in the Din, Oka, and adjoining Volga areas are not clear so far. Corpuses of artefacts and more or less detailed maps are absent. Vladislav Kropotkin s and Anatoliy Ambroz s studies require important amendments, though Michel Kazanski s monograph on the fur trade in East Europe (Казанский 2010) does not investigate mass imports (amphorae, beads, etc.). My present study is based on the analysis of prestigious artefacts, weapons and coins. According to the accounts of Aleksandr Medvedev, Irina Belotserkovskaya, Ol ga Rumyantseva and the author, goods of Greco-Roman origin (mostly beads in the Oka area, and also glass vessels, amphorae and other pottery, some ornaments, coins, etc. on the Don) occurred in those areas in the late Roman period and later. Moreover, these or those imports scattered along the East-European forest-steppe and the south of forest areas forming a bigger concentration in the Chernyakhov culture with its strong connections to the Greco-Roman centres. Since the contacts between different population groups who obtained imported ware certainly existed, these imports were exchanged as well and, in result, their distribution area in Eastern Europe was so wide that they cannot provide direct data concerning the direction of trade routes. In my opinion, indirect accounts are even more important. Trade route should have a system supporting the movement of commodities and human beings. Archaeological materials allow the one to trace where the route began (the area whence the movement started) and intermediate places (where the population that started importing immediately contacted with the receivers). Such mutual contacts necessary resulted in a certain material culture levelling, also reflected in archaeological sources. The focus of this paper is the materials from the upper Don area. Since the discussions in several conferences have shown that the researchers of Central and Western European cultures do not understand the situation in that area in its entirety, I have to make some preliminary notes bringing the colleagues up to date on the case. In the Hunnic period, the sites of the type called Chertovitskoye Zamyatino occurred in the upper Don area (fig. 1). So far archaeological excavations uncovered 21 1 settlements and 121 graves, including three in Zhivotinnoye cemetery, three in Shilovo, 63 in Ksizovo-17, 48 in Ksizovo-19, and single burials in Mukhino-2, Ksizovo-16, Kamenka, and Yasnoye. Especial place in the settlement system in the upper Don area was occupied by sites in Ostraya bend of the Don. It is where the largest settlements from the Hunnic period are located near modern villages of Kamenka, Zamyatino, Ksizovo and Mukhino in the Zadonskiy district of the Lipetsk oblast, i. e. administrative area (Kamenka clus- 1 Every cluster of sites near modern villages of Zamyatino, Kamenka, and Ksizovo is referred to as a single settlement.

283 The Upper Don and the Northern Black Sea Areas ca. 500 AD: Aspects of the Trade Route 283 Fig. 1. Map of the sites in the forest-steppe area on the Don in the Hunnic period: I settlements of Chertovitskoye-Zamyatino type; II inhumation cemeteries and single graves; III stray find; 1 Podgornoye; 2 Starozhivotinnoye-3; 3 Chertovitskoye-3; 4 Chertovitskoye-6; 5 Zamyatino-1, 2, 4, 5, 7 10, 12, 13; 6 Malyy Lipyag, Krutogor ye; 7 Perekhval ; 8 Kamenka-1; 9 Kamenka-4; 10 Kamenka-5; 11 Ksizovo-16, 19, 19A, Ostraya Luka-4; 12 Ksizovo-8; 13 Mukhino-2; 14 Ozerki forest; 15 Kollektiv; 16 Zhivotinnoye; 17 Pekshevo; 18 Nevezhekolodeznoye; 19 Ksizovo-17; 20 Lavy; 21 Ol shanets; 22 Strel bishche-4; 23 Shilovo; 24 Yamnoye; 25 Yablonevo-4

284 284 Andrey OBLOMSKIY ter with unfortified settlements of Kamenka 1, 4, 5 covers ca. 18 hectares, as Igor Biryukov has clarified it in 2010, Zamyatino cluster with unfortified settlements of Zamyatino-1, 2, 4, 5, 7, 8, 10, 12, 13 covers hectares, Mukhino hectares, and Ksizovo with Ksizovo-16, 19, 19А and Ostraya Luka hectares) (Земцов 2003: 108; Бирюков 2004) (fig. 1). The excavations of some of the listed sites discovered remains of bone-carving, bronze-smelting, jewellery, blacksmithing, and ferrous metallurgy. Chain mails were produced or repaired in Zamyatino-5, though the excavations of Ksizovo-19 investigated a metallurgical furnace (Терпиловский 2004: ; Толмачёва 2004; Земцов 2004; Обломский, Усачук 2004). There were many Greco-Roman imports including glass beads and vessels (unfortunately, mostly fragmented), shards of amphorae, wheel-made tableware and lamps, single ornaments and coins. Especially numerous are imported ware from the settlement of Ksizovo-19. The settlements in Ostraya bend of the Don were economical and probably administrative centre of the Upper Don area in the mid-first millennium AD. The Black Sea cultural tradition on the upper Don includes various late antique and barbarian elements. 2 А. Industrial assemblages 2002 excavation of the settlement Ksizovo-19 uncovered two-tier pottery kiln, about 1.6 m high and measuring 1.74 х 2.12 m at the base (fig. 2. 1). The seize of its fire and working chambers was much larger than the Chernyakhov culture kilns (cf.: Бобринский 1991). This kiln was used to fire burnished grey-clay and coarse vessels made with fast-turning wheel (Земцов 2004). The main forms of vessels, such as pots, basin-bowls, and jugs of various forms and shapes (fig. 2), their ornamentation, fabric, and the construction of the kiln meet with analogies at the late antique pottery-making (Обломский, 2004а: ). These vessels sharply differ from hand-made ware of local origin, which predominates in all the sites in the upper Don area. The study of ferrous metal ware from the settlements near Zamyatino village has discovered that they were made with both forest-steppe (from the Chernyakhov and Kiev cultures) and Greco-Roman techniques of metal-working (Толмачёва 2004: ; Терпиловский 2004: ). B. Graves The richest burial in the above-listed group of north-oriented graves was uncovered in the settlement near modern village Mukhino, close to the Ksizovo archaeological complex (Земцов 2003: ) (fig. 3). The noble woman s cloth was embroidered with gold badges, typical of settled population elites in the Pontic area (Казанский, Мастыкова 2009: 242). Direct connection of the deceased lady, or those who buried her, with the lower Don area is indicated by polished ribbed jug within this assemblage (fig. 3. 1), which form and fabric are typical of table ware from Tanais in the Hunnic period (Обломский 2009: ). Two striking assemblages of the Black Sea tradition were investigated in the cemetery of Ksizovo-19, which overlaid settlement layer from the Hunnic period, as described below. Site 1 in excavation trench 7 was a burial of four persons within the same pit to be the unique find in the upper Don area. Grave pit is traced at the bedrock level, below metres-thick humus layer, so the investigations uncovered only the lower construction (fig. 4. 1). The pit of oval plan, m deep below the daily surface, had the top contour measuring 2.6 х 2.2 m and the bottom 2.52 х 2.1 m, and a flat step measuring 0.7 х m on the north. The walls of the pit were charred at their full height (but the step), as it came from reddish-orange traces of burning up to 1 cm thick. There were burnt wooden boards or blocks pressed to the sides of the pit: their traces were discovered as few spots of 1 2-cm-wide narrow strips of charcoal. The pit was filled with humus containing cultural remains related to an earlier settlement (charcoals, plaster, ceramics, and bones of animals). The grave contained four undisturbed skeletons of adults, with well-preserved bones, to form four tiers with the lowest laying on the bedrock bottom of the pit as far from the entrance. i. e. the step, as possible, though the uppermost almost at the step, 15 to 18 cm above the lowermost. Visual observation did not discover traces of fire on the bones. The skeletons were oriented with the heads to the south-west or south-west-south, i. e. to the right of the entrance. Three lower skeletons were extended, the uppermost was crouched on the right side. Above two skeletons, there were single pieces of stone and raw bricks. The burial contained 261 glass beads, predominantly fine red opaque pieces, 2 This division is rather convenient due to active barbarization of the residents of Roman provinces.

285 The Upper Don and the Northern Black Sea Areas ca. 500 AD: Aspects of the Trade Route 285 Fig. 2. Metallurgical furnace and wheel-made grey-clay ware of Ksizovo-19 type (according to Grigoriy Zemtsov and the author). 1 plan and profile of metallurgical furnace in Ksizovo-19 (in squares: 1 plunderers mine; 2 humus with inclusion of cinder and fireplace clay; 3 fired bedrock clay; 4 layer of heavily burned clay with inclusion of straw or grass); 2 4, 6 Ksizovo-19, infill of the metallurgical furnace; 5, 9, 12, Ksizovo-19, settlement; 7 Ksizovo-19, grave 2; 8, 13 Zamyatino-8, settlement; 10 Zamyatino-7, settlement; 11 Zhivotinnoye, grave 4; 2, 4, 7, with polished surface; 3, 5, 8 9, 15 with coarse surface; 6 with sub-polished surface

286 Fig. 3. Artefacts and jug from Mukhino grave. 1 hand-made black-burnished jug; 2 hand-made coarse pot; 3 4 glass speckled beads; 5 mirror (billon); 6 bracelet (silver); 7 lunula pendant (silver); 8 pincers (silver); 9 10, 12 cloth appliqués (gold); 11 bead (gold); 13 toilet set (silver); beads (amber) (drawings by Grigoriy Zemtsov, Igor Gavritukhin, and the author)

287 The Upper Don and the Northern Black Sea Areas ca. 500 AD: Aspects of the Trade Route 287 Fig. 4. Ksizovo-19, site 1 in trench 7. 1 plan and profile of the site; 2 iron cheek-piece

288 288 Andrey OBLOMSKIY Fig. 5. Ksizovo-19, grave 13, plan and profile (henceforth levelling marks are given from the total zero point for uncovered areas)

289 The Upper Don and the Northern Black Sea Areas ca. 500 AD: Aspects of the Trade Route and iron cheek-piece with polyhedral head, two holes, and flat foot (fig. 4. 2). According to Il ya Akhmedov, cheek-pieces with analogous head occurred from the first third or first half of the fifth century. Bronze cheek-pieces of similar form are known among the antiquities synchronous to the second stage of steppe graves from the second half of the fifth century according to Irina Zasetskaya (Ахмедов 1996: 41 42, рис ; Ахмедов 2001: 245; Засецкая 1994: ). The tradition of collective burials in vaults, large pits in ground, and catacombs is well-known in the Crimea in the Roman and the Great Migration periods, both in the cemeteries of Greco-Roman cities and in barbarian part of the peninsula with settled population. These graves differ both in regard to chronology and location (Айбабин 1999: 22 24, 61 66; Храпунов 2004: 148; Айбабин, Хайрединова 2008: 14 18), but the construction of the top part of the Ksizovo grave (vaulting, entryway, etc.) has not been discovered due to the thick humus layer hiding any contour of a pit. Collective burials in large pits and catacombs are known in North Caucasus as well, for example. in the cemetery of Klin-Yar in Kislovodsk hollow (Флёров 2007: ) and, according to Maya Abramova s description, in mountainous areas of North Ossetia (Абрамова 1997: 106). Therefore, since other similar graves are not discovered so far, parallels to site 1 in Ksizovo-19 occur in wide Crimea North Caucasus area. Nevertheless, it is very important that similar grave constructions are not known in the Late Roman and Hunnic period in the upper Don area and its vicinity (forest-steppe Dnieper area and far to the east, on the Oka and middle Volga). Grave 13 was the pit of sub-trapezoid plan, m deep below the daily surface, having the top contour of 2.6 х m and the bottom 2.45 х m (fig. 5). On the floor of the pit, m below the daily surface, there was extended skeleton lying on the back with the head to the north-west, of a gracile woman years old. The bones are generally in poor condition. The head lie on the right with the face to the south. The lower jawbone had teeth up but moved 12 cm to the south-east from the skull. Post-cranial bones remained partly and probably were moved and destroyed by rodents. Bronze non-ornamented bracelet of solid wire was put on the right arm above the hand (fig. 6. 2). Moles moved similar bracelet to the south-eastern area of the pit (fig. 6. 3). In the fill of the pit near the knee of the deceased there was 289 yellow glass eye bead, though clay circular-biconical spindle whorl was located at the wall (fig. 6. 5). The looking over the cultural layer above the pit uncovered bronze finger-ring with spiral scrolls facing in opposite directions (fig. 6. 4). On the left side of the skull, near the eye-socket, there was non-ornamented bronze wire ring with overlapping ends (fig. 6. 1). Near the neck and under the lower jawbone there was an accumulation of glass beads, most likely of fallen down elements of the necklace and head ornament. A part of that head ornament remained in situ below the skull, i. e. on the right of the head (fig. 7. 1). Possible reconstruction of the head ornament consists of a piece of 5-cm-wide ribbon with two lines of blocks of fine yellow and black beads, with fine bronze pendants hanging down (triangular, circular, and in the form of leaves from Greco-Roman funeral wreaths). A thread of bigger beads was fastened to the ribbon (fig. 7. 2), which was locked with a ring with overlapping ends on the forehead. Similar head ornaments are well-known in late antique Europe. Max Martin has used archaeological accounts and images to reconstruct them as ribbons with the lock on the forehead, embroidered with beads or decorated with badges, with loose beaded strings to support hair or veil (Martin 1988: ). An example could be seen on fig. 8 with the heads of Galla Placidia and Amalasunta: although the latter was a daughter to the Ostrogothic king Theodoric the Great, she was portrayed wearing Roman headdress (Мончиньська 2009: рис. 88). C. Hand-made ceramic vessels The sites in the upper Don area contained a small series of polished hand-made vessels of the origin related to the complex of ceramic ware from the late Greco-Roman settlement of Tanais (late fourth and mid-fifth centuries) at the Don estuary. There are bi-conical jugs with flaring neck from a burial near modern Mukhino village (fig. 3. 1), settlement of Kamenka in the Ostraya bend of the Don, and unfortified settlement of Chertovitskoye-6 near modern city of Voronezh (fig. 9. 2, 5), and globular jug with elbow handle from grave 56 in the cemetery of Ksizovo- 17А (fig. 9. 4). Shards of polished hand-made bowls with ribbed decoration at the maximum width of the body originate from settlements of Zamyatino-8, Mukhino-2, and Ksizovo-19 (fig. 9. 1, 3). All the listed forms do not meet with parallels among the table ware from the Kiev cul-

290 290 Andrey OBLOMSKIY Fig. 6. Some artefacts from grave 13 in Ksizovo bronze; 5 clay ture and the Oka origin in the upper Don area, but are frequent amidst polished vessels in Tanais (Обломский 2009: , рис. 3). D. Ornaments Many ornaments and daily life artefacts known in the upper Don area (buckles with trunk-shaped tongues twining around the front side of the frame, mirrors with loop on the back side, bracelets of plain circular rod, conical bells, earrings with polyhedral ends and those in the form of twisted circle, and so on) occurred in the Hunnic period in the northern Black Sea area and more frequently in other European countries. They illustrate the commons style of the period and are not analysed in the given paper. Excavations of Zamyatino-10 discovered buckle tongue and belt appliqué from a Roman metal warrior s belt or an imitations of it (fig ) (Гавритухин 2004). Cultural layer on Ksizovo-19 contained flat silver earring in the form of three joined petals with traces of soldered sockets for stone insets (fig ). Such artefacts were popular in Bosporos, according to Elzara Khairedinova, in the first half of the fifth century AD, though the only find outside the Crimea was from Airan (the Untersiebenbrunn horizon) (Хайрединова 2002: 61, рис ). Zasetskaya has suggested a wider chronology of the earrings from the Kerch vaults, that is the second half of the fourth and the first half of the fifth centuries (Засецкая 1993: 53, 77, табл ; ).

291 The Upper Don and the Northern Black Sea Areas ca. 500 AD: Aspects of the Trade Route 291 Fig. 7. Ksizovo-19, grave 13. Location of ornaments below the skull (1), reconstructed section of head ribbon (2) There are 29 two-piece fibulae of bronze and iron, with solid catch-plate or returned foot from the upper Don sites from the Hunnic period. Although different classification systems attribute these brooches to different taxons due to specific details, 3 yet they applied the same technology of fasten- 3 Anatoliy Ambroz s group 15, series VI ( Lebyazh ye ), variant 4; group 16, sub-group 3, series I; and group 17, sub-group 1, variant II (Амброз 1966); or Anna Mastykova s five series within the group of arched fibulae (Мастыкова 2009: 41 46); however, Igor Gavritukhin has suggested three series of Bosporan fibulae with wide ring to hold the spring (Гавритухин 2008: ).

292 292 Andrey OBLOMSKIY Fig. 8. Portraits of noble ladies from the Great Migration period. 1 Galla Placidia; 2 Amalasunta (based on Max Martin and Magdalena Mączińska)

293 The Upper Don and the Northern Black Sea Areas ca. 500 AD: Aspects of the Trade Route 293 Fig. 9. Hand-made polished ware of Tanais origin from the upper Don area. 1 Mukhino-2; 2 Chertovitskoye-6; 3 Zamyatino-8; 4 Ksizovo-17А; 5 Kamenka-4 (drawings by Igor Biryukov, Grigoriy Zemtsov, and the author) ing the spring axle with its rod passing through wide ring or socket formed by top edge of the bow curved up or down (fig ). The main distribution area of these fibulae in Eastern Europe is the northern Black Sea coast, the lower Don and the North Caucasus, within the period from the late fourth to the sixth century, and as late as the seventh century in the North Caucasus (Обломский 2009: ). Such a high concentration of these fibulae has not been documented in the forest-steppe area other than the upper Don. Some fibulae from the forest-steppe Don area have bronze cylinders put on the spring axle (for example: Обломский 2009: рис ), and the same decoration occurs on the spring axle ends of few fibulae from Tanais, probably indicating the single tradition of craftsmanship (Гавритухин, Воронцов 2008: 42). Specific group of ornaments consists of radiateheaded brooches, chip-carved brooches, and flyshaped brooches and appliqués.

294 294 Andrey OBLOMSKIY Fig. 10. Several ornaments and belt fittings from Ostraya bend of the Don. 1 silver; 2, 3, 14 iron; 4 11 bronze; 1, 4 12 Ksizovo-19; 2 3 Zamyatino-10 (drawings by Igor Gavritukhin) Bronze radiate-headed brooch with pentagonal head-plate from Ksizovo-19 (fig ) has prototypes in the Danube area, particularly in the cemetery of Untersiebenbrunn (Tejral 1997: Abb , 7). Igor Gavritukhin has attributed the small cast brooches of similar style to the type Vyshkov- Chersonesos, appeared as an imitation of large artefacts with pentagonal and triangular head-plates. The pentagonal head-plate relates the Ksizovo brooch to the Pontic series, with its general chronology in the middle and the second half of the fifth century (Гавритухин 2000: ). Concerning the style, this group also includes the second bronze radiate-headed brooch from Ksizovo-19 (fig ). Mastykova has related fragment of a very similar item with the foot-plate forked below discovered in Krasnodar water storage with the Danube tradition (Мастыкова 2009: рис ). Two solid-cast chip-carved (Kerbschnitt) fibulae originate from Ksizovo-8 and Ksizovo-19 and three fragments of the foot-plates of similar artefacts from Ksizovo-19 (fig. 12), all belonging to Gavritukhin s group Levice-Tokari. Semicircular

295 The Upper Don and the Northern Black Sea Areas ca. 500 AD: Aspects of the Trade Route 295 Fig. 11. Ornaments from Ksizovo-19. 1, 2, 4 7 bronze; 3 silver panel above the spring of complete brooch is a feature of the researcher s Black Sea series. The closest analogy to the Ksizovo-8 brooch is the find from Tokari on the left bank of the Dnieper (Гавритухин 1994: 33, 37; рис. 2. 7). Apart from the latter brooch, all others in this series mentioned by Gavritukhin occurred in the Black Sea area, mostly in the Crimea. Jaroslav Tejral has cited examples of such clasps with semicircular head-plate from the Danube area, for example in Záhony, Hungary (Tejral 1997: Abb ). The fly brooch from Ksizovo-19 is made of bronze, though the appliqué is of silver (fig ). Its typical features are three wings of subtriangular section, pronounced neck, and schematic head of the insect. A very close analogy to the appliqué is a find vault no. 14/1914 in Chersonesos (Гавритухин, Казанский 2006: рис ; Айбабин 1990: рис ). There is complete and very similar brooch but with extension above (Амброз 1966: табл ). The Ksizovo-19 brooch is almost the same as the one from vault

296 296 Andrey OBLOMSKIY Fig. 12. Bronze brooches. 1 Ksizovo-8; 2 4 Ksizovo-19; 5 Ol shanets 181/1902 in Kerch (Амброз 1966: табл ; Айбабин 1990: рис ). Aleksandr Aibabin has dated Crimean fly brooches from the second half of the fifth to the first half of the sixth century (Айбабин 1990: 26; рис ). Similar ornaments occured in the Middle Danube area (Vinski 1957: Abb. 5. 5, 16). The publication of the cemetery of Viminacium attributes such clasp to the type of faceted cicada brooches (fibules-cigales á facettes) with a note that in the Danube area and in the Crimea brooches of the type date back from the second third of the fifth to the early sixth century (Ivanisević et al. 2006: 17; fig ). There is bronze fly brooch with two wings with their ends curved up and long neck from Ol shanets on the Snova river (stray find) (fig ), of the North Caucasus origin. According to Gavritukhin and Kazanski, this brooch combines features of variants 3 and 4, though Mastykova has attributed it to type 1 (Гавритухин, Казанский 2006: 328; Мастыкова 2009: 40). In the Caucasus, the general date for artefacts of variants 3 and 4 is from the second half of the fifth to the first half of the seventh centuries (Гавритухин, Казанский 2006: ). There are one complete and four fragmented bracelets with parallel concentric circles in relief on the ends (two of them were secondly used as rings) from Ksizovo-19 (cultural layer in excavation trenches, chance finds) (fig ). Another bracelet of the type was discovered in the cemetery of Zamyatino-7 (Обломский 2004c: рис ). Although similar ornaments are typical of the North Caucasus, Ciscaucasia, and the Crimea (Мастыкова 2009: 67), they are also known in the Danube area (Шаламон, Баркоци 1982: рис ).

297 The Upper Don and the Northern Black Sea Areas ca. 500 AD: Aspects of the Trade Route Bronze bracelet from grave 5 in the cemetery of Zhivotinnoye, with narrow rod of round cross-section, rectangular panel, and ornament of concentric circles on its ends, has direct analogies in the late Tanais (Обломский 2004b: 224; рис , 19). Bracelets with very similar ornamentation originate from the cemetery of Dyurso (Мастыкова 2009: рис ). There is a specimen with the ends of similar shape but without ornamentation discovered in Ksizovo-19. E. House-building Irina Zin kovskaya has suggested that the construction of building 1 in the Third Chertovitskoye fortified settlement followed the tradition of the late antique Tanais (Зиньковская 2008: 22), where houses with two and more rooms sunken into soil were popular in the Hunnic period (Ulrich 1999: ), some with dome kilns. The building in the Third Chertovitskoye had stone walls, measured 3 x 4 m, was divided into two parts by a partition, and was sunken into ground ( carved into the south-eastern slope of the inner rampart, in the publication s words) (Медведев 1998: 50). There also was wattle-and-daub kiln. Although the house in Chertovitskoye-3 actually reminds the buildings of Tanais, yet Medvedev and Denis Akimov suggest another parallel, the construction at the rampart of the Chernyakhov settlement of Bashmachka (Медведев 1998: 56). Both suggestions have the right to exist, but the dome kiln in Chertovitskoye makes Zin kovskaya s idea more probable. * * * The data cited above supply evidences that in the upper Don area, and primarily in the Ostraya bend of the Don, there were migrants from the south (north-eastern Black Sea area in wide sense of the term, including the Crimea, North Caucasus, and lower Don): craftsmen (a potter, blacksmiths, possibly jewellers), some horse-riders (cheek-piece in the collective grave in Ksizovo-19 most likely symbolised of the dead man s occupation), and noblemen (Mukhino). The contacts of the upper Don and Tanais populations are more understandable because of hand-made vessels, house-building, and some typological details of fibulae. In the upper 297 Don area, there was a fashion of ornaments from the Black Sea and North Caucasus, when fibulae with curved bow and returned foot and those with solid catch-plate were numerous, and the production of other types of clasps in the Pontic area started under the influence of Germanic traditions from the Danube. Obviously, the upper Don artefacts typologically similar to the Pontic ware appeared there in relation to the migration of some population. It should be mentioned that the upper Don area was a link in the connection between the Black Sea Caucasus and the Oka area. Although prior to the publication of full catalogues we can hardly evaluate the distribution of southern artefacts in the Oka basin in the Hunnic period and in the early Middle Ages, yet in Ryazan, Oka, and ancient Mordovia cemeteries there were well-known bracelets with slightly flattened anthropo- and zoomorphic ends, rather often discovered in the Crimea and the North Caucasus (Мастыкова 2009: 67 69) and with the roots in the Pontic area (Труфанов 2001: 71). Belotserkovskaya has pointed out that such bracelets appeared in the barrow culture of Ryazan Oka area in the period 3, from the fifth century (possibly without its first decades) to the early sixth century (Вихляев et al. 2008: 33, рис ; Белоцерковская 2007: 195, ; рис , 15; , 4 5). Similar distribution axis (the Oka and middle Volga areas the North Caucasus) is reflected by bracelets with the ends forming triangular blades, decorated with lines of pointed design (Вихляев et. al. 2008: 33, рис ; Белоцерковская 2007, рис ; Мастыкова 2009: 69 70). Artefacts of both types are known in the sites in the Ostraya bend of the Don. 4 The list of parallels would possibly be continued after new publications of artefacts, primarily from the Oka area. In the forest-steppe area on the Don, within the polyethnic population of the mid-first millennium AD, there were migrants from the Oka area, from the areas of Moshchino, Late D yakovo, and Ryazan Oka cultures (according to specific hand-made vessels and some types of ornaments), discussed in other publications of mine. Therefore, the contacts of the Black Sea Caucasian and the Oka populations were direct in the Upper Don area. Apart from the fashion of some ornaments mov- 4 Besides two finds listed above (Обломский 2009: 264), there is another fragment of bracelet with triangular blades discovered in the settlement of Mukhino-2 in 2010, though seven fragments of bronze bracelets with anthropo- and zoomorphic ends originate from Ksizovo-19.

298 298 Andrey OBLOMSKIY ing from south to the north, there also were some other goods migrating in reverse direction (pectoral ornament from the Oka in the cemetery of Tanais and set of northern ornaments excavated by Pavel Leont yev in the settlement of Tanais in the collection of the State Historical Museum) (Обломский 2009: с. 265), 5 though in a much smaller scale. Hand-made vessels with analogies in the Kiev culture site was possibly brought to Tanais in the same way (Обломский 2010а: 64 65, рис ). On the south of this route there was Tanais with its undoubted connections with the upper Don population. Late antique settlement was renovated in the Hunnic period, i. e. in the period of appearance of ethnically mixed populations in the upper Don area, at place of the city that had perished in the mid-third century conflagration. This regular settlement was built according to the plan developed in advance, with division into upper and lower riverside area, the latter probably at the port. The population of this town was mixed, with its main bulk migrated from far-away north-western Black Sea, i. e. from the southern Chernyakhov area. There are almost no traces of mass marketoriented industries in the settlement, so they even did not have their own wheel-made vessels (Обломский 2010а). Late antique Tanais surprieses the researcher with the huge number of imported vessels, primarily amphorae, though red slip pottery was also rather popular. According to Sergey Bezuglov, there is relatively big number of Roman and Bosporan coins in Tanais (Безуглов 2001). All these data makes me to suppose late antique Tanais as a port and obviously a trading centre within the Hunnic realm, and a possible starting point of the trade route upstream the Don, as it comes from the contacts between the populations of Tanais and the upper Don area. Therefore, we can see a certain route connecting the northern Black Sea area with the Oka basin (including neighbouring areas), with various types of contacts of populations recorded along its line in the mid-first millennium AD. The intermediary link in this chain was the upper Don area (primarily the settlement cluster in the Ostraya bend of the Don, a kind of capital of this area) where lived both the migrants from the Oka and the Pontic areas (in their wide historical-geographical frames). When did this trade route function? The analysis of datable artefacts suggests that the main period of the late antique settlement in Tanais was the late fourth and the first half of the fifth centuries. Some artefacts like the matrix to make badges of Shipovo style witness for the life in this settlement continued in the second half of the fifth century (Обломский 2010b). The sites of Chertovitskoye-Zamyatino type in the Don area and synchronous inhumation graves date back to the late fourth and fifth centuries (Обломский 2007: 76 77). Obviously, the trade route from the Black Sea to the Oka area existed during that chronological period. Bibliography Ivanisević V., Kazanski M., Mastykova A. Les nécropoles de Viminacium à l époque des Grandes migrations. Paris, Martin M. Grabfunde des 6. Jahrhunderts aus der Kirche St. Peter und Paul in Mels SG. // Archäologie der Schweiz Teiral J. Neue Aspecte der frühvolkerwanderungszitlichen Chronologie im Mitteldonaurum // Neue Beitrage zur Erforschung der Spätantike im mittleren Donauraum. Brno, Ulrich M. Tanais Eine antike Stadt am Rand der Steppe (Die spätantike Phase) // Arbeitsblätter für Restauratoren. 1. Mainz, Vinski Z. Zikadenschmuck aus Jugoslavien // Jahrbuch der RGZM Абрамова М. П. Ранние аланы Северного Кавказа III V вв. н. э. М., Айбабин А. И. Хронология могильников Крыма позднеримского и раннесредневекового времени // МАИЭТ I. 5 I am indebted to Akhmedov for the information concerning Leont yev finds: he prepares them to publication in cooperation with Irina Tolochko.

299 The Upper Don and the Northern Black Sea Areas ca. 500 AD: Aspects of the Trade Route 299 Айбабин А. И. Этническая история ранневизантийского Крыма. Симферополь Айбабин А. И., Хайрединова Э. А. Могильник у с. Лучистое. T. 1. Симферополь; Керчь, Амброз А. К. Фибулы юга Европейской части СССР II в. до н. э. IV в. н. э. / САИ. Д М., Ахмедов И. Р. Удила // Гавритухин И. О., Обломский А. М. Гапоновский клад и его культурноисторический контекст (РСМ. 3). М., Ахмедов И. Р. Коллекция предметов конского убора из Керчи // Поздние скифы Крыма / ТГИМ М., Безуглов С. И. Денежное обращение Танаиса (III в. до н. э. V в. н. э.). Дисс. канд. ист. наук. М., 2001 (Archive of the Institute of Archaeology of the Russian Academy of Sciences, collection R 2, no. 2654). Белоцерковская И. В. Инвентарь женских погребений // Восточная Европа в середине 1 тыс. н. э. / РСМ. 9. М., Бирюков И. Е. Острая Лука Дона в гуннское время. Общая карта памятников // Острая Лука Дона в древности. Замятинский археологический комплекс гуннского времени / РСМ. 6. М., Бобринский А. А. Гончарные мастерские и горны Восточной Европы. М., Вихляев В. И, Беговаткин А. А., Зеленцова О. В., Шитов В. Н. Хронология могильников населе ния I XIV вв. западной части Среднего Поволжья. Саранск, Гавритухин И. О. Причерноморская серия фибул группы Левице-Токари (к изучению восточногерманского культурного наследия) // БСб. 4. М., Гавритухин И. О. Финал традиций культур римского времени в Восточном Прикарпатье // Die spätrömische Kaiserzeit und die frühe Völkerwanderungszeit in Mittel- und Osteuropa. Łodź, Гавритухин И. О. Приложение к главе 1. Детали поясов «позднеримского воинского» стиля с поселения Замятино-10 // Острая Лука Дона в древности. Замятинский археологический комплекс гуннского времени / РСМ. 6. М., Гавритухин И. О. Фибула из раскопок А. Г. Атавина в Фанагории в 1989 г. // Древности Юга России (памяти А. Г. Атавина). М., Гавритухин И. О., Воронцов А. М. Фибулы верхнеокско-донского водораздела: двучленные прогнутые подвязные и со сплошным приемником // Лесная и лесостепная зоны Восточной Европы в эпохи римский влияний и Великого переселения народов. 1. Тула, Гавритухин И. О., Казанский М. М. Боспор, тетракситы и Северный Кавказ во второй половине V VI вв. // АВ. 13. СПб., Засецкая И. П. Материалы Боспорского некрополя второй половины IV первой половины V вв. н. э. // МАИЭТ III. Засецкая И. П. Культура кочевников южнорусских степей в гуннскую эпоху (конец IV V вв.). СПб., Земцов Г. Л. Миграционные потоки III V вв. и верхнедонской регион (на примере поселения Мухино-2) // Контактные зоны Евразии на рубеже эпох. Самара, Земцов Г. Л. Гончарный горн на поселении Ксизово-19 // Острая Лука Дона в древности. Замятинский археологический комплекс гуннского времени / РСМ. 6. М., Зиньковская И. В. Археологические памятники раннегуннского времени Верхнего и Нижнего Дона (сравнительный анализ) // Восточнославянский мир Днепро-Донского междуречья и кочевники южно-русских степей в эпоху раннего средневековья. Воронеж, Казанский М. М. Скандинавская меховая торговля и «Восточный путь» в эпоху переселения на родов // Stratum plus

300 300 Andrey OBLOMSKIY Казанский М. М., Мастыкова А. В. Кочевые и оседлые варвары в Восточной Европе в гуннское время // Дивногорский сборник. 1. Воронеж, Мастыкова А. В. Женский костюм Центрального и Западного Предкавказья в конце IV середине VI в. М., Медведев А. П. III Чертовицкое городище (материалы 1-ой половины 1 тыс. н. э.). // Археологические памятники Верхнего Подонья первой половины 1 тыс. н. э. Воронеж, Мончиньська М. Велике переселення народів. Історія неспокійної епохи ІV та V століть. Київ, Обломский А. М. Замятинский археологический комплекс «перекрёсток этнокультурных традиций» // Острая Лука Дона в древности. Замятинский археологический комплекс гун нского времени / РСМ. 6. М., 2004а. Обломский А. М. Раннесредневековое трупоположение у с. Лихачёвка Полтавской обл. // Культурные трансформации и взаимовлияния в Днепровском регионе на исходе римского времени и в раннем средневековье. СПб., 2004b. Обломский А. М. Поселение Замятино-7 // Острая Лука Дона в древности. Замятинский археологический комплекс гуннского времени / РСМ. 6. М., 2004c. Обломский А. М. Танаис и Верхнее Подонье в гуннскую эпоху (проблема контактов населения) // Дивногорский сборник. 1. Воронеж, Обломский А. М. Этнокультурные компоненты населения поселка гуннского времени в Танаисе // Черняхівська культура. Актуальні проблеми досліджень (до 40-річчя археологічної експедиції НПУ ім. М. П. Драгоманова) (Тези доповідей). Київ, 2010а. Обломский А. М. Хронология поселения Танаис позднеантичного периода // Лесная и лесостепная зоны Восточной Европы в эпохи римских влияний и Великого переселения народов. Конференция 2/1. Тула, 2010b. Обломский А. М., Усачук А. Н. Технология изготовления гребней в Замятино-5 и Замятино-8 и мастерские гребенщиков // Острая Лука Дона в древности. Замятинский археологический комплекс гуннского времени / РСМ. 6. М., Терпиловский Р. В. Социально-экономический аспект исследования материалов Замятинского археологического комплекса //Острая Лука Дона в древности. Замятинский археологический комплекс гуннского времени / РСМ. 6. М., Толмачёва М. М. Результаты металлографического изучения кузнечных изделий памятников Замятинского археологического комплекса // Острая Лука Дона в древности. Замятинский археологический комплекс гуннского времени / РСМ. 6. М., Труфанов А. А. К вопросу о хронологии браслетов с зооморфными окончаниями (по материалам крымских могильников позднескифского времени) // Поздние скифы Крыма / ТГИМ М., Флёров В. С. Постпогребальные обряды Центрального Предкавказья в I в. до н. э. IV в. н. э. и Восточной Европы в IV в. до н. э. XIV в. н. э. М., Хайрединова Э. А. Женский костюм варваров Юго-Западного Крыма V первой половины VI в. // МАИЭТ IX. Храпунов И. Н. Этническая история Крыма в раннем железном веке / БИ VI. Шаламон А., Баркоци Л. Археологические данные к периодизации позднеримской Паннонии ( гг.) // Древности эпохи Великого переселения народов V VIII веков. М., 1982.

301 The Upper Don and the Northern Black Sea Areas ca. 500 AD: Aspects of the Trade Route 301 Андрей ОБЛОМСКИЙ Верхнее Подонье и Северное Причерноморье в середине I тысячелетия н. э. Проблема торгового пути Резюме Статья посвящена обзору археологических материалов, которые позволяют реконстру и- ровать путь из Азовского моря вверх по Дону в Поочье в гуннское время (конец IV V вв.). На археологических материалах могут быть прослежены начало пути, промежуточные пунк ты, где происходил непосредственный контакт населения территории, откуда исходил импорт, и его получателей. В результате взаим ных контактов обязательно происходит некото рая ни велировка материальной культуры. В Верхнем Подонье в конце IV V в. известны поселения типа Чертовицкого-Замятино и могильники круга Животинного (рис. 1). Осо бое место в структуре заселения верхнедонского региона занимали поселки на Острой луке Дона. Именно здесь, около сёл Каменка, Замятино, Ксизово и Мухино Задонского района Липецкой области расположены самые обширные по площади поселения гуннского периода. На них зафиксированы остатки косторезного, бронзолитейного, ювелирного, куз нечного, гончарного ремесел, черной метал лургии. В Замятино-5 изготавливали или ремонтировали кольчуги. Достаточно часты предметы античного импорта (бусы, обломки стеклянных сосудов, амфор, гончарных столовых сосудов, светильников, отдельные украшения и монеты). Поселки на Острой луке Дона были экономическим и, вероятно, административным центром верхнедонского региона в середине I тыс. н. э. Имеются также данные о непосредственном присутствии причерноморского населения на территории донской лесостепи, при этом причерноморская традиция в Верхнем Подонье представлена как позднеантичными, так и варварскими культурными элементами. А. Ремесленные комплексы и традиции. На поселении Ксизово-19 раскопан гончарный горн двухъярусной конструкции. В печи обжигали сероглиняную лощеную и шероховатую ке рамику, изготовленную на круге быстрого вращения (рис. 2). В результате специального исследования изделий из черного метал ла поселений у села Замятино выяснилось, что при изготовлении использовались не толь ко лесостепные (черняховские и киевские), но и позднеантичные приемы металлообработки. Б. Погребения. Наиболее богатое захоронение обнаружено на поселении Мухино (рис. 3). Расшитый золотыми бляшками костюм погребенной здесь знатной женщины по М. М. Казанскому и А. В. Мастыковой харак терен для аристократии оседлого понтий ского населения. Два ярких комплекса при черноморской традиции исследованы на мо гильнике Ксизово-19. Объект 1 раскопа 7 (рис. 4) представляет собой погребение четырех человек в одной яме. Традиция коллективных захоронений в склепах, больших грунтовых ямах и катакомбах в III V вв. хорошо известна в Крыму и на Северном Кавказе. В погребении 13 захоронена женщина возрастом года. Под черепом in situ cохранилась часть украшения головы, которую можно реконструировать, как отрезок ленты шириной около 5 см, на которую в два ряда были нашиты блоки желтого и черного бисера, со свисавшими с них маленькими бронзовыми подвесками. К ленте была прицеплена низка более крупных бусин. На лбу она замыкалась бронзовым кольцом (рис. 5 7). Подобные украшения головы по М. Мартину хорошо известны в позднеантичной Европе (рис. 8). В. Лепная керамика. На памятниках Верхнего Подонья имеется небольшая серия лепных лощеных сосудов, которые связаны по происхождению с керамическим комплексом позднеантичного поселения в Танаисе (конец IV середина V в.), расположенного в устье Дона (рис. 9). Г. Украшения. В Замятино-10 обнаружены детали металлической гарнитуры римского воинского пояса или подражающие им (по И. О. Гавритухину) (рис , 3). Из культурного слоя Ксизово-19 происходит се ребряная пластинчатая серьга со следами гнёзд

302 302 Andrey OBLOMSKIY для каменных вставок (рис ), по Э. А. Хайрединовой характерная для Боспора первой половины V в.. На памятниках Верхнего Подонья найдено 29 фибул из бронзы и железа с прогнутым корпусом и широким кольцом или муфтой для крепления оси пружины (рис ). Основной ареал этих фибул в Восточной Европе Северное Причерноморье, низовья Дона и Северный Кавказ, суммарная дата конец IV VI вв. Бронзовая двупластинчатая фибула из Ксизово-19 (рис ) относится к понтийской серии типа «Вышков-Херсонес» по И. О. Гавритухину, общая дата которой середина вторая половина V в. Целая литая фибула с кербшнитным орнаментом из Ксизово-8 (рис ) относится к причерноморской серии группы «Левице-Токари» по И. О. Гавритухину. На Острой луке До на известны еще одна целая фибула и об ломки подобных изделий (рис ). В Ксизово-19 найдены фибула-«цикада» и похожая накладка (рис , 4). В Причерноморье А. И. Айбабин датирует такие вещи второй половиной V в. первой половиной VI в. Похожие украшения известны в Среднем Подунавье. Фи була-«цикада» из Ольшанца (рис ) име ет северокавказское происхождение. По И. О. Гав ритухину и М. М. Ка занскому, близкие изделия датируются вто рой половиной V первой половиной VII в. На памятниках Острой луки Дона найдено шесть обломков браслетов с рельефным орнаментом в виде параллельных концентрических кругов на концах (рис ). По А. В. Мастыковой, подобные украшения характерны для Северного Кавказа, Закавказья, Крыма, хотя встречаются и в Подунавье. Два бронзовых браслета с прямоугольным щитком и узким круглым в сечении стержнем из Животинного могильника и Ксизово-19 имеют прямые аналогии в позднем Танаисе и на могильнике Дюрсо. Д. Домостроительство. На Третьем Чертовицком городище исследована постройка размером 3 х 4 м, с каменными стенами и глинобитной печью, разделённая перегородкой на 2 части и углубленная в грунт. По планировке, конструкции стен и печи она аналогична домам позднего горизонта Танаиса. Приведенные выше данные свидетельствуют о присутствии в Верхнем Подонье выходцев из Северо-Восточного Причерноморья, включая Крым, Северный Кавказ, низовья Дона. Более определенны связи населения Верхнего Подонья и Танаиса. В Верхнем Подонье распространяется мода на причерноморскосеверокавказские украшения. Некоторые типы браслетов демонстрируют распространение близких украшений по оси Северное Причерноморье Верхнее Подонье Поочье в гуннское время и в начале средневековья. Таким образом, вырисовывается некий путь, связывающий Северное Причерноморье и бассейн Оки (включая окрестные территории), по линии которого в середине I тыс. н. э. фиксируются различные по характеру контакты населения. Начальным звеном этого пути был позднеантичный Танаис, промежуточным Верхнее Подонье. Статья подготовлена при поддержке программы фундаментальных исследований Секции истории РАН «Нации и государство в мировой истории».

303 303 Radosław PROCHOWICZ Carnelian Beads in the Cemetery of the Wielbark Culture at Szelków Nowy, Poland Souvenirs from the Black Sea? On the maps illustrating the situation during Roman and Pre-Roman Iron Age the lower Orzyc region was until recent considered as an empty area (cf.: Godłowski 1985: maps 1 6, 9; Dąbrowska, Liana 1986: map 1; Andrzejowski 2001: fig. 2, 3, 11 13). Excavations and a series of accidental discoveries in the past 20 years revealed in this area the existence of at least nine settlements and cemeteries located in close neighbourhood. Materials derived from these, which can be attributed to the Przeworsk and Wielbark cultures and dated from the younger Pre-Roman Iron Age till the Early Migration period, in most cases do not differ from those discovered in other sites in this part of the region. In addition, there are also objects the souvenirs from around the world mentioned in the title which are a manifestation of far-range contacts of the population from the lower Orzyc area. Polyhedral beads made of chalcedony are examples for such a category of items. Two of them were found in Przeradowo and 14 more come from the cemetery at Szelków Nowy. In the first of these sites, the objects we are interested in were discovered in a brown-black layer of sand at a depth of 0.25 to 0.4 m from the ground surface. Objects exposed in their vicinity, including those made of copper alloy: a fragment of a brooch type Gródek 47, variant II (Andrzejowski, Prochowicz, Rakowski 2008: 55, fig. 4m) or a damaged beak-shaped strap end (Prochowicz 2005), allow one to assign the beads to the Wielbark culture and dating to the phase C3 D. Determination of raw material from which they have been made is currently quite difficult due to the partial burn and associated discoloration of the subjects. The context of discovering the polyhedral beads from Szelków Nowy seems to be much more interesting they come from inhumations in graves 130 and 133. The first of the graves was discovered at a depth of about 0.3 m from the ground surface as a rectangular dark brown discoloration with dimensions of 1.05 x 1.5 m, oriented along the NW-SE axis (fig. 1). In the profile the grave pit outline was similar to an inverted trapeze and about 0.8 m thick. The upper parts of the fill were covered with the dark brown layer 0.4 m thick containing scattered, probably blown, very large stones and two glass beads: one complete and one melted. Below this layer there was a light brown filling. Approximately 0.55 m from the observed top level of the object pit dimensions decreased slightly and amounted 0.5 x 1.3 m. At that depth two pieces of the pottery put on the side were discovered, belonging to the types R Ib and R XVIIIA 1 (fig , 24). Between them, in the middle of their height, a piece of a mandible and several teeth of a six months old child 2 were found. Slightly lower, covered with some unspecified organic matter, two silver brooches type A 172 were discovered (fig , 20), between them polyhedral beads of chalcedony type TM 499 (fig ). Another brooch type A 172 made of copper alloy was found few centimetres to the south of them (fig. 1, 2. 21). Next to it there were about nine strongly weathered, crumbling glass beads (probably dark blue). Two copper alloy lunulae and two (?) lines of glass beads (types TM 2, TM 30a, TM 30/47, TM 57, TM 126, TM 330), amber beads (type TM 389, TM 402 ) and pendants in the shape of ring with silver wire (type Beckman 15) were 1 In this article, I have used the following typologies of findings identified in the text with abbreviations: A (Almgren 1923) for the brooches, TM (Tempelmann-Mączyńska 1985) and G (Гопкало 2008) for the beads, ML (Madyda-Legutko 1987) for the buckles, and R (Wołągiewicz 1993) for the pottery. 2 The age was determined by Łukasz Maurycy Stanaszek of the Laboratory of Anthropology at the State Archaeological Museum in Warsaw.

304 304 Radosław PROCHOWICZ Fig. 1. Szelków Nowy (district Maków Mazowiecki, Poland), grave amber; silver; glass. (Drawing: Bartłomiej Karch, the author; photo: Agata Wiśniewska)

305 Carnelian Beads in the Cemetery of the Wielbark Culture at Szelków Nowy, Poland Souvenirs from the Black Sea? found directly in the south of the brooch (fig ; ). Among these items there was also a buckle with a thickened oval frame made of copper alloy, type ML AH12 (fig ). Underneath the brooches and beads remains of wood were preserved, and in between also pieces of two cloths or leather and textile (Maik 2012: 51, fig. 77, 78). Object 133 was observed at a depth of about 0.4 m as an irregular brownish-blackish discoloration in the outline similar to the oval elongated in the NS axis with dimensions of 0.6 x 1 m. In the NE corner there were two probably burnt stones (fig. 3). Approximately 20 cm below the outline of the grave pit became sharper similar to the oval with dimensions of 0.75 x 1.3 m. At this level in the northern part of the pit two pottery vessels, types R Ic i R XIVB (fig. 3. 5, 6; in the bigger of them a polyhedral glass bead type G XVII/12), and in the vicinity six glass beads (type TM 30, TM 175 and G XVII/12), three amber beads (type TM 389, TM 402), two chalcedony beads (type TM 499), and a copper alloy brooch were discovered (fig. 3. 3, 4). In the central and southern part of the pit (fig. 3. 1, 2, 4; 4) there were 15 more beads made of glass (type TM 14, TM 24, TM 30/47, TM 172, TM 362e), amber (type TM 389, TM 430, TM 440), and chalcedony (type TM 499), two copper alloy brooches, similar to the type A 185, arranged heads toward the top of the object and copper alloy buckle with buckle plate (type AH 16). The grave pit has in the profile an outline similar to a rectangle and thickness of about 0.3 m. The filling was patchy, light brown, in the northern part of the top darker with single fragments of burned bones and pottery. Throughout the object filling single minor coals were found. The basis for the dating of the first of the mentioned graves is three brooches type A 172. The researchers do not agree as to their chronology. Some attribute them to phase C2 C3 (e. g. Schuster 2004: 88 89), others to phase C3 (e. g. Kokowski 1995: 15, 44, 48, fig. 2r; Niezabitowska-Wiśniewska 2007: 645, ), and some to phases C3 D (Woźniak 2011: 183). Resolving that issue goes far beyond the scope of this study, nevertheless it seems that the 305 pieces from Szelków should be placed in the last of the above mentioned compartments. This is indicated by both dating of accompanying buckle type ML AH 12 (Madyda-Legutko 1987: 64) and beads type TM 330 (Tempelmann-Mączyńska 1985: 57). Other items of the grave equipment such as the beads types TM 2, TM 30 and TM 30/47 are assigned to much wider chronological framework, and thus not very useful for our discussion. The dating of the Grave 133 is a much more complex problem. The brooches discovered in it are relatively rare (cf. Schuster 2004: 98, fig. 2; Woźniak 2011: 184) and require an extensive discussion. Two of them came from the same mold as evidenced by the soft spot on the outside of the catchplate (fig. 3. 1, 2). Clear, indelible traces file (?) visible on the inside surface of the bows of the both pieces and lack of the knob on the head of one of the brooches can be explained by the fact that the objects were not completed. The closest analogy to our brooches is the one from grave 708 of Kleszewo, 3 district Pułtusk as well as the one found stray in the cemetery in Niedanowo (Ziemlińska- Odojowa 1999: 109, pl. CCXXIX. 14), district Nidzica and the brooch from grave 252 in Stary Targ 4, district Malbork. Due to their form and the presence of the spring terminals and head knobs they refer to the type A 185, except the piece from Stary Targ. The lack of the frame-like foot in the case of Szelków finds may be caused by the fact they were not completed. However, on the basis of a few well-dated copies of brooches A 185, the findings from Szelków can be applied to the phase C2 C3 of the Roman period (Schuster 2004: 85 88, with further literature ). Unlike these two brooches, the third one (fig. 3. 3) shows signs of usage like the smoothness inside the bow and the visible wear of the pin at the point of contact with the catchplate. In the available literature, I could not find an accurate analogy for it. The territorially nearest copies of similarly shaped bow, foot and head are among the so-called Raupenfibeln (Tuszyńska 1988). Although the brooch of Szelków is lacking the characteristic decoration of the bow, 3 Unpublished material from the collections of the Museum of Ancient Metallurgy in Pruszkow prepared by Marcin Woźniak, whom I thank very much for the information. 4 Due to the complete catch-plate, this brooch was classified by Aleksandra Żórawska as type A 170 (Żórawska 2005: 123, pl. XXXI / ). However the find was lost during the World War II and now only a sketch by Waldemar Haym is available. On this drawing, above the passage of the bow in the foot several cross bars are envisaged. In a situation when the brooch has a full catch-plate, it is difficult to consider this as a wrap sticking the catch-plate to the bow. Perhaps it is the ornament imitating brooch item, present at brooches of Szelków and Kleszewo.

306 306 Radosław PROCHOWICZ Fig. 2. Szelków Nowy (district Maków Mazowiecki, Poland), grave carnelian; 7 16 glass; 17, 18, 21, 22 copper alloy; 19, 20 silver; 23, 24 clay. (Drawing: Bartłomiej Karch, Anna Potoczny; photo: the author) the delicate incision may be some form of the caterpillar ornament imitation. It can be assumed that for the unique piece of Szelków we may adopt the general dating of the Raupenfibeln included in phases C2 D. The two graves of Szelków can be placed in the advanced phase C of the Roman period, and maybe even in the beginning of the Migration Period. Not only their dating is similar. Both of them have almost identical dimensions, numerous references in the assemblage like the number of brooches, chalcedony beads of type TM 499 and glass beads of type G XVII/12, and placing of the two vessels smaller and bigger in the northern part of the grave

307 Carnelian Beads in the Cemetery of the Wielbark Culture at Szelków Nowy, Poland Souvenirs from the Black Sea? 307 Fig. 3. Szelków Nowy (district Maków Mazowiecki, Poland), grave copper alloy; 5, 6 clay; a carnelian bead; b glass bead; c amber bead; d brooch; e buckle. (Drawing: Bartłomiej Karch, the author; photo: the author) pit. In both cases, one could see traces of secondary pits containing burnt stones or finds. It appears that individuals buried in Graves 130 and 133 had some relationship (family?) and their burial was either at the same time, or in a small time interval. Very interesting is also the fact that in many respects they stand out against other similar burials of the Wielbark culture. The first one to mention is the depositing of three brooches in each of the graves in Szelków. In the Wielbark culture in the advanced phase C of the Roman period, the dead were buried typically with two or one brooch, and very often with none (Tempelmann-Mączyńska 1989: 73). The burials with three brooches as in Pruszcz Gdański (Schindler 1941: 53 54, pl. 4), district Gdańsk, Weklice, district Elbląg (graves 34, 174, 220, 275, 288, 331) (Natuniewicz-Sekuła, Okulicz-Kozaryn 2011: 32 33, 60, 69, 79 80, 83, 91, 108, pl. XV, XIV, LXXI, LXXII, XCII, CXVI CXIX, CXXVIII, CXLV, CLXXXVIII), Bialowieża (Dzierżykraj- Rogalski, Jaskanis 1961), district Hajnówka are exceptional. It is interesting that except the grave of Bialowieża, all the rest are located in the region of the mouth of the Vistula, and the finds of Szelków are the only so dated in Mazovia. 5 5 Magdalena Tempelmann-Mączyńska in her work dedicated to the outfit in middle European Barbaricum mentions two graves with three brooches from the outside of the mouth of the Vistula (Tempelmann-Mączyńska 1989: 73). These are the Białowieża grave and grave 9 of Stara Wieś. In the latter case there had to be some mistake made, because in that burial only two brooches were found (cf. Radig 1942: , fig. 2. 9, 13, pl. II. c, d).

308 308 Radosław PROCHOWICZ The order of brooches similar to the one observed in grave 133 is rare, too. 6 Normally, in burials with three brooches, they are being discovered in the northern part of the pit, where, according to the Wielbark custom was the head of the deceased, next to the shoulder and arm bones (if preserved; e. g. Weklice, graves 174, 220, 275). In Szelków in that part of the pit a single brooch was found, and two other along with the buckle in the middle part. Lack of bones and not very long pit (up to 1.3 m in length), can be the basis for treating the grave as symbolic and finds distribution as quite arbitrary. Perhaps, however, the deceased was buried in grave 133, but then due to some unknown reason excavated and displacements of the remains along with the objects located on them were made. The evidence for the re-opening of the grave after the funeral may be the different colour of the filling in the central and northern parts of the upper part of the filling, and several beads located to the north from the pottery they could have been displaced there when digging or filling. It is possible that despite the re-opening of the grave, the deceased was left in peace, in that case the order of the brooches reflects the way the shroud or the outfit of the buried person were wrapped. The question of determining the age of the person buried in grave 133 remains. Taking into consideration the distance between the brooches from the northern and central part of the pit, of about 40 cm, it seems that it was an adult. While taking into account the size of the grave pit had to be buried with flexed legs otherwise simply would not fit. 7 Of course, the presented reconstruction is fairly hypothetical but because of the reported cases of burials with flexed legs (Tempelmann-Mączyńska 1992: 197) of the Wielbark cemeteries, it should be considered as probable. For some other reasons grave 130 in which an infant was buried, should be considered as an exceptional. In the materials of the Wielbark Culture burials of children under one year of age are not often reported. In general, the assemblage was quite simple (Skóra 2012: 143), and even if it contained neither items made of precious metals nor other precious gifts (Skóra, Troszczyńska-Antosik 2012: 154). Studies of Kalina Skóra devoted to children in the Wielbark culture indicate that the funerary equipment of the dead at early stage of life should rather be interpreted in terms of the emotional gifts rather than emphasizing the social status. In this perspective, the rich arrangement discovered in grave 130, should be considered as a proof of an extremely strong bond which united the buried child with parents or family. In graves 130 and 133, in spite the very exceptional elements of the funerary rite, attention should be also paid to some of the items discovered in them. These are in particular the glass beads of type TM 30/47, TM 172 and TM 175b, TM 330, TM 362e and G XVII/12 and especially interesting for us stone beads of type TM 499 (fig. 4). I have included to the first of these varieties the dark blue beads made of clear glass, which in terms of the general form resemble copies of type TM 30, however, in contrast to them have a very wide opening like beads type TM 47 (fig. 2.7, 8, 13, 16, ). Beads of this kind are indeed included in the Wielbark culture materials, but rather as a manifestation of far-reaching influence from the area of the Chernyakhov (Sîntana de Mureş) culture (Kokowski 1995: 54, fig. 73 / map 52). 8 Two beads type TM 330 (fig. 2. 9, 15) and beads type G XVII/12 should be considered as imports from the South-East. The first of them, except the discovery of Jaskinia Ciemna in Ojców (Mączyńska 1970: 20, pl ), district Krakow and grave 57 of Kutowa, district Hajnówka, site II (Jaskanis 2012; 169, fig. 21, pl. 86/57. 1), has no parallels in the areas of Central Barbaricum, but on the northern shores of the Black Sea beads of that kind are quite numerous (Алексеева 1978: таб as type 325; Гопкало 2008: 56, таб. VI/X. 8, карта 28 as type G X/8). Slightly more complex is the issue of the beads type G XVII/12 (fig , 15; ), which are in the form of cubes with bevelled corners (Гопкало 2008: 41 42, таб. IV/XVII. 12). Thus, they should not be included in the type TM 126, including longitudinal polyhedral beads made of transparent blue glass (Tempelmann-Mączyńska 1985: 19, 37 38, pl ). However, the author herself did not apply this classification criterion rig- 6 The deployment of brooches the same as in the Grave 133 was founded only in grave 288 of Weklice (Natuniewicz-Sekula, Okulicz-Kozaryn 2011: 83, pl. CXXVIII) among the materials of developed Wielbark culture of the phase C and the beginning of the Migration period. 7 On the subject of grave pits that seem too small compared to the growth of the person buried in it see: Tempelmann-Mączyńska 1992: In that study beads type TM 30/47 are described under the name of type P5.

309 Carnelian Beads in the Cemetery of the Wielbark Culture at Szelków Nowy, Poland Souvenirs from the Black Sea? 309 Fig. 4. Szelków Nowy (district Maków Mazowiecki, Poland), grave amber; 7 10, glass; carnelian. (Drawing: Bartłomiej Karch) orously, resulting including in the type TM 126 or generally in group XIV (Tempelmann-Mączyńska 1985: ) specimens not having a longitudinal form. 9 Nevertheless, it seems that such cases were not too numerous 10 and beads type G XVII/12 discovered in the Central European Barbaricum should be regarded as a manifestation of contacts with the northern shores of the Black Sea, where they were quite common element of outfits. The origin of bead types TM 172, TM 175b (fig ) and TM 362 (fig ) is problematic. They are known from quite a few and scattered finds. The nearest analogy to beads type 172 and 175b were discovered in the Wielbark culture cemetery in Brudnice. 11 Further copies of beads similar both in form and type of glass, are being discovered in present-day Germany. 12 Individual discoveries of four-clover-shaped beads, but unlike the ones we are interested in, made of clear glass, come from the area of the Masłomęcz 13 and the Dębczyno 14 group. Beads type TM 362 are known mostly from the present-day Germany, and in the Wielbark culture area from Malbork-Wielbark, district Malbork, Kleszewo, district Pułtusk, and Cecele, district Siemiatycze (Tempelmann-Mączyńska 1985: , map 54, where further literature). 9 Including: Dębczyno, district Białogard, cemetery: graves 1, 9 (dark red colour; Machajewski 1993: 14, pl. II/1. 6, 7; IV/9.9) and settlement: feature 300 (colour unknown: Machajewski 1992: pl. XXII. 6), Borkowice, district Koszalin (former Borkenhagen, Kreis Köslin), grave III (dark blue colour), Redło, district Świdwin (former Redel, Kreis Belgrade), stray finds from graves (pink-red colour), Strammnica, district Kolobrzeg (former Alt-Tramm, Kreis Kolberg), stray finds (dark blue colour?; Eggers, Stary 2001: 148, 151, 154, pl ; ; ; ; ). 10 The fact that polyhedral beads, similar to the form of cubes are not too frequent finds in the Central European Barbaricum area can be proofed by the summary of cuboctahedral beads in the study of Erich Blume (1915: ), in which only a few are described as shaped like a dice German Würfel. 11 Unpublished studies of Andrzej Szela from the Warsaw University Institute of Archaeology, whom I thank for the information. 12 E. g. Großweißandt, Landkreis Köthen, Grave 5, Merseburg Kiesgrube Rosener, site 8, Landkreis Merseburg- Querfurt, grave 43, Weißenfels Beudefeld site 27, Landkreis Weißenfels, stray finds (Schmidt, Bemman 2008: 51 52, 88 89, 122, pl. 47/49/ ; 2. 10; 103/75/43. 1, 4; ), Gerlachsheim, Landkreis Baden-Württembergs, grave 2 (Dauber 1958: 141, pl. A ), and Neuhof, Landkreises Märkisch-Oderland, stray finds from graves and Pritzier, Landkreis Ludwigslust-Parchim, stray finds (Tempelmann-Mączyńska 1985: 177, 242, 250, with further literature). 13 Grave 108, Gródek on the Bug, district Hrubieszów (Kokowski 1993/I: 81; 1993/II, fig. 99b). 14 Dębczyno, site 3, feature 300, district Białograd (Machajewski 1992: 51 52, pl. XXII. 5, 12).

310 310 Radosław PROCHOWICZ Against that background, issues related to polyhedral stone beads seems to be interesting. They are generally made of chalcedony, 15 also, as in the case of Szelków finds of its red and brown varieties carnelian. In the Central European Barbaricum (fig. 5) the most numerous finds of these objects come from the area of the Przeworsk culture (about 150 pieces from 19 sites) 16 and the Wielbark culture (about 110 pieces from 25 sites 17 ). All of them are regarded as a manifestation of contacts with the 15 In those cases when type of material was specified. Very often, in the description of this type of beads the type of stone, or even colour were not given. 16 1) Biejków distr. Białobrzegi, site 9, grave 1 four pieces (Szela 2007: 286, 288, fig ); 2) Chmielów Piaskowy, district Ostrowiec Świętokrzyski, grave 4 four pieces, grave 63 two pieces (Godłowski, Wichman 1998: 16, 42, 58, pl. V/4.4; LXXXI/63.9, 10); 3) Ciebłowice Duże, district Tomaszów Mazowiecki, site 1, grave 32 one piece (Dzięgielewska, Kulczyńska: 19, 53, pl. XXIV/32.2; XCIX.9); 4) Drochlin, district Czestochowa, stray finds three pieces (Kaczanowski 1987: 49, 78, pl. XXVII.39; XXVIII.38, 41); 5) Kietrz, district Glubczyce, grave 1704 one piece (Gedl 1972: , fig. 5e); 6) Kocierzew Południowy, district Łowicz, site 3, grave 5 one piece (Skowron 1998: 80, pl. IX. 6); 7) Komorów, district Rawa Mazowiecka stray finds 19 pieces (Wawrzeniecki 1908: 63; 1908b: 90; 1903: 39, pl. X. 13, 14); 8) Kuców, district Belchatow, grave pieces (Olędzki 1985: 57, , pl. XII. 4, 7); 9) Mokra, district Kłobuck, grave 255 one piece (Biborski 2002: 230, fig. 2e); 10) Olsztyn, district Czestochowa cemetery with cremation layer one piece (Szydłowski 1974: 151, pl. CLXXXIg ); 11) Opatów, district Kłobuck, grave pieces (Madyda-Legutko, Rodzińska-Nowak, Zagórska-Telega 2011/I: 226; 2011/II: pl. CCCXLIX ); 12) Rajbrot, district Bochnia, unspecified one piece (Biborski 2002: 90 91, fig. 11); 13) Rybno, district Kłobuck, grave 5 15 pieces (Godłowski 1966: 151, fig ); 14) Spicymierz, district Poddębice, grave 261 three pieces (Kietlińska, Dabrowska 1963: 180, 192, pl. XXXV. 27); 15) Starachowice, grave VI 21 pieces (Jamka 1959, p , fig. 11g); 16) Wiktorów, district Zgierz, grave 7 one piece, grave 16 four pieces, grave 28 eight pieces (Moszczyński 2010: 102, 109, 123, fig ; ; ); 17); Zadowice, district Kalisz, grave 112 three pieces, grave 151 four pieces, grave 417 one piece, grave 421 one piece, grave 429? pieces, grave 519 one piece (Abramowicz Lepówna 1957: 31, pl. XX ; Tempelmann-Mączyńska 1985: 324; Kaszewska 1988: 57); 18) Żdżarów, district Sochaczew, grave 73 one piece, grave 123 one piece, grave 143 two pieces, stray finds three pieces (Nowakowski 2003: 292, 297, 299, 306, pl. XVII/73. 1; XXXI/123. 4; XXXVI/143. 2, 3; LV. 9 11); 40) Białobrzegi, distr. Łańcut, site 18, stray finds one piece (Ligoda, Podgórska-Czopek 2011: 143) ) Brudnice, district Żuromin, grave 279 five pieces (unpublished studies of Andrzej Szela from the Warsaw University Institute of Archaeology); 20) Cecele, district Siemiatycze, grave 219 one piece, grave 285 four pieces, grave 371 one piece, grave 390 one piece, grave 403 two pieces, grave 499 one piece, grave 504 three pieces, grave 511 one piece, grave 518 ten pieces, stray finds one piece (Jaskanis 1996: 43, 70, 52, 54, 56, 67 69, 82, 106, pl. XXIX/219. 4; XXXVI/ ; XLIII/371. 2; XLVI/390. 2; L. 22, 23; LX/499. 2; LXI/ ; LXIV/511. 2; LXV. 8 17; LXXIX. 17); 21) Frombork, district Braniewo, stray finds (Peiser, Kemke 1914); 22) Goździk, district Garwolin, barrow 1 one piece (Niewęgłowski 1984: 137, 149, fig. 7. 2); 23) Gródek on the Bug, district Hrubieszów, st. 1A, stray finds one piece (Gładysz, Kokowski 2002: 285, pl. VI. 4); Gródek on the Bug, district Hrubieszów, st. 1C, grave 64 one piece (Kokowski 1993/I: 51; 1993/II: fig. 55i); 24) Janowo, district Brodnica, grave 7 one piece (Hahuła 1993: 146, 152, pl. III. 6); 25) Kiełpino, district Kartuzy, treasure two pieces (La Baume 1925: 22, fig. 7g); 26) Kleszewo, district Pułtusk, grave 708 seven pieces, stray finds one piece (unpublished material from the collections of the Museum of Ancient Metallurgy in Pruszkow prepared by Marcin Wozniak, whom I thank very much for the information; cf. Tempelmann-Mączyńska 1985: 285); 27) Kutowa, district Hajnówka, site II, Graves 64 four pieces, 67 one piece (Jaskanis 2012: 170, fig. 20, pl. 87/ ; 87/67. 1); 28) Masłomęcz, district Hrubieszów, site 15, grave 120? pieces, grave 163? pieces, grave 244? pieces, cremation layer ten pieces (Gładysz, Kokowski 2002: 285); 29) Niedanowo, district Nidzica, grave 142 five pieces, gra ve 325 one piece, grave 391 one piece, grave 512 one piece, stray finds one piece (Ziemlińska-Odojowa 1999: 33, 64, 73, 90, 110; pl. XL/142. 4; CXIX. 6; CXL/391. 4; CLXXXVI/512. 4; CCCXXX. 71); 30) Przeradowo, district Maków Mazowiecki, stray finds two pieces (unpublished collection of State Archaeological Museum in Warsaw [PMA/IV/10372]; cf. Prochowicz 2006); 31) Rostołty, district Białystok, barrow 2 two pieces (Jaskanis 2012: 17, fig. 19, pl , 7); 32) Skorbicze, Białoruś, stray finds five pieces (Andrzejowski 2005: 29, pl. 5w); 33) Sławka Wielka/Schläfken, district Nidzica, Grave VII three pieces (Peiser 1909: 329, 331, fig. 197); 34) Stara Wieś, district Siedlce, Gra ve 63 one piece (Dąbrowska 1973: 493, pl. IV. 11); 35) Swaryczów, district Zamość, st. 1, grave 20/11 one piece (Sadowski 2008: 306, fig. 7. 5); 36) Szelków Nowy, district Maków Mazowiecki, grave 130 six pieces, grave 133 eight pieces (unpublished materials of the State Archaeological Museum in Warsaw [PMA/IV/10472]); 37) Szpaki, district Bielsk Podlaski, grave 5 two pieces (Rusin 2008: 298, fig. 5. 9); 38) Ulów, district Tomaszów Lubelski, stray finds two pieces (Niezabitowska-Wiśniewska 2007: 21, fig. 18); 39) Warszawa-Kawęczyn, district Warszawa, grave 15 one piece, grave 16 one piece, stray finds five pieces (Jankowska 1962: 334, 336, 346; pl. LXVII.19; LXXII ); 41) Modła, distr. Mława, Grave 190 five pieces, Grave 237 one piece (Andrzejowski 2006: 25, fig. 12/190. 1; 12/237. 5); 42) Nur- Kolonia, distr. Ostrów Mazowiecka, stray finds 1 piece (unpublished collection of State Archaeological Museum in Warsaw [PMA/IV/5772:221]); 43) Wielka Kletna, Białowieża National Park, Grave 1 one, stray finds two pieces (Krasnodębski, Dulinicz, Samojlik, Olczak, Jędzejewska, 2008: , 370, list 2. 5, 8, fig. 4. 4; 10. 3).

311 Carnelian Beads in the Cemetery of the Wielbark Culture at Szelków Nowy, Poland Souvenirs from the Black Sea? 311 Fig. 5. Finds of carnelian (A) and other semi-precious stone (B) beads from the Przeworsk and Wielbark culture areas. Numbers correspond to the list of finds (see footnotes 43 and 44). (Drawing: the author) population of the Carpathian Basin area and the Black Sea coast, where they were one of the key elements of the outfit. This hypothesis is entitled also by the studies on production sites of these beads and raw material sourcing. In many older studies both beads and raw material for their production were considered as imports from North Africa or the Middle or Far East(e. g. Алексеева 1982: 5 6). In more recent literature, some argue that they might have been produced in the Carpathian Basin and in the areas of the Chernyakhov culture. Workshops, producing beads of stone discovered Tibiscum a town located next to the legion camp in Dacia (Benea

312 312 Radosław PROCHOWICZ 2004: ), are an evidence for that. Another proof is a lump of untreated carnelian (dimensions 7 x 4.5 x 3 cm) discovered in the Chernyakhov culture cemetery of Balcaty I (Балцаты I). According to the author of the proceedings on the material discovered on that site, this object might have come from the destroyed grave of a jeweller (Фёдоров 1960: 225, 239). Regarding the sites for the manufacture of semi-precious stone beads there are no researches and analyses available confirming their exploitation in the Roman period, except the indications of potential deposits in the Carpathian Basin (e. g. Vaday 1989: ; Benea 2004: 107). Extremely interesting is as well the identification of such contacts, which led to the emergence of the stone beads in areas north of the Carpathians. In my attempts to resolve that issue for obvious reasons I will focus on the Wielbark materials. The first stone beads throughout the settlements occupied by population of this culture occurred during B2/C1 phase 18, so at around the same time as the beginning of the Wielbark expansion toward south-east, where the fashion for this type of decoration was growing. 19 It is interesting that in all the Wielbark objects, which are dated to the beginning of phase C of the Roman period or generally to this phase, one up to three beads of that kind were found. It looks as if they were an insignificant element of outfit, and treated as a curiosity, gadget, souvenir. It is also possible that the colour and foreign origin caused treating those beads as talismans or amulets. The situation changes in the graves dated to the C3 D phase 20. Then appeared the whole sets of beads, counting five or more units, which form if visible necklaces. So probably we are dealing with a more complete adaptation of the patterns coming from the South-East. The question is whether they reached the area occupied by the population of the Wielbark culture only as objects with information about their purpose, or maybe together with the people from the South, for example of the Chernyakhov culture area. Talking about the finds from the Szelków Nowy I would tempt for the latter possibility. As I tried to show above, graves 130 and 133 discovered in Szelków are quite unique when compared to other similar burials dated to the Wielbark culture. Any deviation in the funeral rite and assemblage used to be explained by the status of the dead in the local community or the unusual circumstances of their death. The accumulation of objects mainly beads which have the most numerous analogies either in materials of the Chernyakhov culture, or more broadly in the Black Sea area, shows that at least one adult a woman (?) from grave 133 might come from this region. Of course it is difficult to decide on the origin in the case of a child under one year of age from grave 130. However, many similarities of the both objects suggest some kind of relationship between the people buried in them perhaps relationship of mother and child. Under this assumption, the various differences associated with these burials can be explained by the origin of the dead and their incomplete assimilation to the new environment. The fact that both of them were at least partly assimilated is proofed by the local elements of the assemblage: brooches and pottery. An additional impulse for the specific treatment of both of the deceased might have been caused by the reason of their death which is impossible to determine on the basis of the remains. This interpretation accepted for the objects from Szelków Nowy, cannot be automatically imposed on all the stone beads from the area of the Wielbark culture. Every single case requires a separate analysis. But regardless of the way the beads arrived, they were always a souvenir from afar, often from the Black Sea. 18 Janowo, grave 7, district Brodnica (Hahuła 1993: 146, 152, pl. III. 6). 19 According to Adrea Vaday earliest polyhedral stone beads appear in the Carpathian Basin in the second century BC and their number is growing, to reach its peak in the last decades of the fourth century BC (Vaday 1989: ). 20 E. g. Cecele, graves 285, 518, district Siemiatycze (Jaskanis 1996: 43, 69, 106; pl. XXXVI/ ; LXIV/511. 2), Niedanowo, grave 142, district Nidzica (Ziemlińska-Odojowa 1999: 33, pl. XL/142. 4), Szelków Nowy, graves 130, 133.

313 Carnelian Beads in the Cemetery of the Wielbark Culture at Szelków Nowy, Poland Souvenirs from the Black Sea? 313 Bibliography Abramowicz A., Lepówna B. Materiały z cmentarzyska w Zadowicach, pow. Kalisz, cz. II // Prace i Materiały Muzeum Archeologicznego i Etnograficznego w Łodzi. Seria archelogiczna. 2. Łódź, Almgren O. Studien über nordauropäische Fibelformen der ersten nachchristlichen Jahrhunderte mit Berücksichtigung der provinzialrömischen und südrussischen Formen // Mannus-Bibliothek. 32. Leipzig, Andrzejowski J. Przemiany osadnicze i kulturowe na wschodnim Mazowszu i południowym Podlasiu u schyłku starożytności // Najstarsze dzieje Podlasia w świetle źródeł archeologicznych. Siedlce, Andrzejowski J. Skorbicze, d. gm, Kosicze, pow. Brześć nad Bugiem (pow. Brześć Litewski, gubernia grodzieńska // Zabytki z okresu wpływów rzymskich, średniowiecza i czasów nowożytnych z Białorusi w zbiorach Państwowego Muzeum Archeologicznego w Warszawie. Warszawa, Andrzejowski J. Nekropola ze schyłku starożytności w Modle koło Mławy // Pogranicze trzech światów. Kontakty kultur przeworskiej, wielbarskiej I bogaczewskiej w świetle materiałów z badań I poszukiwań archeologicznych / Swiatowit. Supplement Series P. Prehistory and Middle Age. XIV. Warszawa, Andrzejowski J., Prochowicz R., Rakowski T. Brooches Type Gródek 47 and Their Dating // The Turbulent Epoch. New materials from the Late Roman Period and The Migration Period. 1 / Monumenta Studia Gothica. V/1). Lublin, Benea D. Die römischen Perlenwerkstätten aus Tibiscum. Timisoara, Biborski M. Ratownicze badania wykopaliskowe na cmentarzysku kutlry przeworskiej z późnego okresu wpływów rzymskich i wczesnej fazy wędrówek ludów w Mokrej, woj. Śląskie // Badania archeologiczne na Górnym Śląsku i ziemiach pogranicznych w latach Katowice, Blume E. Die germanischen Stämme und die Kulturen zwischen Oder und Passarge zur römischen Kaiserzeit. Teil II: Material // Mannus-Bibliothek. 14. Würzburg, Dąbrowska T. Nowe materiały z cmentarzyska w Starej Wsi, pow. Węgrów // Wiadomości Archeologiczne. XXXVII / 4. Warszawa, Dąbrowska T., Liana T. Stan i potrzeby badań nad młodszym okresem przedrzymskim i okresem wpływów rzymskich na Mazowszu // Stan i potrzeby badań nad młodszym okresem przedrzymskim i okresem wpływów rzymskich w Polsce. Kraków, Dauber A. Neue Funde der Völkerwanderungszeit aus Baden (Gerlachsheim, Ilresheim, Zeutern) // Badische Fundberichte. 21. Freiburg; Karsruhe, Dzierżykraj-Rogalski T., Jaskanis J. Grób szkieletowy dziecka z późnego okresu rzymskiego odkryty w 1959 roku w Białowieży, pow. Hajnówka // Rocznik Białostocki. I. Białystok, Dzięgielewska M., Kulczyńska M. Ciebłowice Duże. Ein Gräberfeld der Przeworsk-Kultur im südwestlichen Masowien // Monumenta Archaeologica Barbarica. XIV. Warszawa, Eggers H. J., Stary P. F. Funde der vorrömischen Eisenzeit, der römischen Kaiserzeit und der Völkerwanderungszeit in Pömmern // Beiträge zur Ur- und Frühgeschichte Mecklenburg-Vorpommerns. 18. Lübstorf, Gedl B. Szkieletowy grób z okresu wędrówek ludów na stanowisku 1 w Kietrzu, pow. Głubczyce // Materiały Archeologiczne. XIII. Kraków, Gładysz M., Kokowski A. Gródek nad Bugiem cmentarzysko ludności grupy masłomęckiej z Zamczyska // Varia Barbarica, Zenoni Woźniak ab amicis dicata / Monumenta Archaeologica Barbarica, Series Gemina. I. Warszawa; Lublin, Godłowski K. Badania wykopaliskowe na cmentarzyskach z okresu wpływów rzymskich w Walenczowie i Rybnie, pow. Kłobuck, w 1964 roku // Sprawozdania Archeologiczne. XVIII. Wrocław, Godłowski K. Przemiany kulturowe i osadnicze w południowej i środkowej Polsce w młodszym okresie przedrzymskim i w okresie rzymskim. Wrocław, Godłowski K., Wichman T. Chmielów Piaskowy. Ein Gräberfeld der Przeworsk-Kultur im Świętokrzyskie-Gebirge // Monumenta Archaeologica Barbarica. VI. Kraków, Hahuła K. Cmentarzysko kultury wielbarskiej w Janowie, woj. toruńskie // Wiadomości Archeologiczne. LII/2 ( ). Warszawa, Jankowska B. Cmentarzysko z późnego okresu rzymskiego w Warszawie-Kawęczynie // Wiadomości Archeologiczne. XXVIII/4. Warszawa, 1962.

314 314 Radosław PROCHOWICZ Jaskanis J. Cecele. Ein Gräberfeld der Przeworsk- und Wielbark-Kultur in Ostpolen // Monumenta Archaeolo gica Barbarica. II. Kraków, Jaskanis J. Wodzowskie kurhany kultury wielbarskiej na Podlasiu. Białystok, Kaczanowski P. Drochlin ciałopalne cmentarzysko kultury przeworskiej w okresu wpływów rzymskich // Zeszyty Naukowe Uniwersytetu Jagiellońskiego. DCCCXXVIII. Prace Archeologiczne. 40. Kraków, Kaszewska E. Wstępne wyniki badań cmentarzyska z okresu przedrzymskiego i rzymskiego w Zadowicach pod Kaliszem // Prace i Materiały Muzeum Archeologicznego i Etnograficznego w Łodzi. Seria archeologiczna. 32. Łódź, Kietlińska A., Dąbrowska T. Cmentarzysko z okresu wpływów rzymskich we wsi Spicymierz, powiat Turek // Materiały Starożytne. IX. Wrocław; Warszawa; Kraków, Kokowski A. Gródek nad Bugiem. Cmentarzysko grupy masłomęckiej, część I III // Lubelskie Materiały Archeologiczne. VII/1 3. Lublin, Kokowski A. Grupa masłomęcka. Z badań nad przemianami kultury gotów w młodszym okresie rzymskim. Lublin, Krasnodębski D., Dulinicz M., Samojlik T., Olczak H., Jędzejewska B. Cmentarzysko ciałopalne kultury wielbarskiej w uroczysku Wielka Kletna (Białowieski Park Narodowy, woj. podlaskie) // Wiadomości Archeologiczne. LX. Warszawa, La Baume W. Germanische Funde der Völkerwanderungszeit aus Nordostdeutschland // Blätter für deutsche Vorgeschichte. 31. Danzig, Ligoda J., Podgórska-Czopek J. Autostradą w przeszłość. Katalog wystawy. Rzeszów, Machajewski H. Z badań nad chronologią dębczyńskiej grupy kulturowej w dorzeczu Parsęty // Uniwersytet im. Adama Mickiewicza w Poznaniu. Seria Archeologiczna. 38. Poznań, Machajewski H. Materiały do badań nad obrządkiem pogrzebowym ludności kultury dębczyńskiej // Uniwersytet im. Adama Mickiewicza w Poznaniu. Seria Archeologiczna. 39. Poznań, Mączyńska M. Materiały z okresu wpływów rzymskich z Jaskini Ciemnej w Ojcowie, pow. Olkusz // Materiały Archeologiczne. XI. Kraków, Madyda-Legutko R. Die Gürtelschnallen der römischen Kaiserzeit und der frühen Völkerwanderungszeit im mitteleuropäischen Barbaricum / BAR IS Oxford, Madyda-Legutko R., Rodzińska-Nowak M., Zagórska-Telega J. Opatów Fpl. 1. Ein Gräberfeld der Przeworsk-Kultur im nordwestlichen Kleinpolen // Monumenta Archeologica Barbarica. XV/1 2. Warszawa; Kraków, Maik J. Włókiennictwo kultury wielbarskiej. Łódź, Moszczyński J. Zespół osadniczy kultury przeworskiej na stanowisku 1 4 w Wiktrorowie, pow. Zgierz, woj. łódzkie. Ratownicze badania archeologiczne na trasie autostrady A-2 w woj. Łódzkim // Via Archaeologica Lodziensis. II. Łódź, Natuniewicz-Sekuła M. Okulicz-Kozaryn J. A Cemetery of the Wielbark Culture on the Eastern Margin of Vistula Delta (Excavation ) // Monumenta Archaeologica Barbarica. XVII. Warszawa, Niewęgłowski A. Dwa Cmentarzyska z okresu rzymskiego w Goździku, gm. Borowie, woj. Siedleckie // Sprawozdania Archeologiczne. XXXV (1983). Wrocław; Warszawa; Kraków; Gdańsk; Łódź, Niezabitowska-Wiśniewska B. Ulów tajemnica starożytnego Roztocza, siódma wystawa Instytutu Archeologii Uniwerstytetu Marii Curie-Skłodowskiej w Muzeum UMCS. Lublin, Nowakowski Z. Cmentarzysko kultury przeworskiej w Żdżarowie, pow. Sochaczewski // Wiadomości Archeologiczne. LVI. Warszawa, Peiser F. E. Schläfken, Kr. Neidenburg // Sitzungsberichte der Altertumsgesellschaft Prussia. 22 ( ). Königsberg, Peiser F. E., Kemke H. Der Depotfund von Frauenburg // Sitzungsberichte der Altertumsgesellschaft Prussia. 23/1 ( ). Königsberg, Pogranicze trzech światów. Kontakty kultur przeworskiej, wielbarskiej i bogaczewskiej w świetle materiałów z badań i poszukiwań archeologicznych // Światowit, Supplement Series P: Prehistory and Middle Ages. XIV. Warszawa, Prochowicz R. Znalezisko dziobowatego okucia końca pasa z Przeradowa, pow. Makowski // Wiadomości Archeologiczne. LVI. Warszawa, Prochowicz R. Osada w Przeradowie // Archeologia Żywa. 4/34 (2005/2006). Warszawa, 2006.

315 Carnelian Beads in the Cemetery of the Wielbark Culture at Szelków Nowy, Poland Souvenirs from the Black Sea? 315 Radig W. Das ostgermanische Gräberfeld von Stara Wies, Kreis Sokolow // Die Burg. 3/2. Krakau, Rusin K. Grave of the Wielbark Culture from the Younger Roman Period under Barrow no. 1 in Szpaki, Wyszki Commune, Bielsk Podlaski District, Podlasie Voivodship // The Turbulent Epoch. New Materials from the Late Roman Period and Migration Period. I / Monumenta Studia Gothica. V/1. Lublin, Sadowski S. The Youngest Phase of Wielbark Culture Cemetery in Swaryczów, Site. 1 (Komarów-Osada Commune, Zamość District, Lublin Province) // The Turbulent Epoch. New materials from the Late Roman Period and Migration Period. II / Monumenta Studia Gothica. V/2. Lublin, Schindler R. Ein Grabfund der Völkerwanderungszeit aus Praußt // Gothiskandza. 3. Danzig, Schmidt B., Bemann J. Körperbestattungen der jüngeren Römischen Kaiserzeit und Völkerwanderungszeit Mitteldeutschlands. Katalog // Veröffentlichungen des Landesamtes für Denkmalpflege und Archäologie Sachsen-Anhalt Landesmuseum für Vorgeschichte. 61. Halle (Saale), Schuster J. Untersuchungen zu den spätkaiserzeitlichen Fibelformen Almgren 185 und 172 und deren gegenseitigen Verhältnis // Veröffentlichungen zur brandenburgischen Landesarchäologie. 35 (2001). Potsdam, Skowron J. Materiały ze zniszczonego cmentarzyska kultury przeworskiej w Kocierzewie Południowym, pow. Łowicz, woj. łódzkie, na stanowisku 3 // Łódzkie Sprawozdania Archeologiczne. IV. Łódź, Skóra K. Groby dziecięce w kulturze wielbarskiej // Światowit, fasc. B. VII/XLVIII ( ). Warszawa, Skóra K., Troszyńska-Antosik M. Cmentarzysko z okresu rzymskiego w Waplewie, pow. Оlsztyński // Wiadomości Archeologiczne. LXIII. Warszawa, Szela A. Nietypowy zespół grobowy z cmentarzyska kultury przeworskiej w Biejkowie, stan. 9, pow. Białobrzegi // Okres rzymski i wędrówek ludów w północnej Polsce i materiały z konferencji Cmentarzyska z okresu rzymskiego jako miejsca kultu Białe Błota, maja 2000 r. / Acta Universitatis Lodziensis. Folia Archaeologica. 25. Łódź, Szydłowski J. Trzy cmentarzyska typu dobrodzieńskiego // Rocznik Muzeum Górnośląskiego w Bytomiu. Archeologia Bytom. 11. Bytom, Tempelmann-Mączyńska M. Das Frauentrachtzubehör des mittel- und osteuropäischen Barbaricums in der römischen Kaiserzeit. Kraków, Tempelmann-Mączyńska M. Die Perlen der römischen Kaiserzeit und der frühen Phase der Völkerwanderungszeit im mitteleuropäischen Barbaricum // Römisch-Germanische Forschungen. 43. Mainz am Rhein, Tempelmann-Mączyńska M. Specyficzne pochówki kultry wielbarskiej // Folia Archaeologica. 16. Łódź, Vaday A. Die sarmatischen Denkmäler des Komitats Szolnok. Ein Beitrag zur Archäologie und geschichte des sarmatischen Barbaricums // Antaeus. Communicationes ex Instituto Archaeologico Akademiae Scientiarum Hungaricae ( ). Budapest, Wawrzeniecki M. Poszukiwania archeologiczne w Królestwie Polskiem // Materyały Antropologiczno-Archeologiczne i Etnograficzne. VI. Kraków, Wawrzeniecki M. Poszukiwania archeologiczne w Królestwie Polskiem dokonane w lipcu, sierpniu i wrześniu 1904 roku przez Maryana Wawrzenieckiego i Szczęsnego Jastrzębowskiego // Materyały Antropologiczno- Archeologiczne i Etnograficzne. X. Kraków, Wawrzeniecki M. Poszukiwania archeologiczne w Królestwie Polskiem dokonane przez Maryana Wawrzenieckiego 1905 i 1906 roku // Materyały Antropologiczno-Archeologiczne i Etnograficzne. X. Kraków, Wołągiewicz R. Ceramika kultury wielbarskiej między Bałtykiem a Morzem Czarnym. Szczecin, Woźniak M. Cmentarzysko kultury wielbarskiej w Kozłówku, pow. nidzicki (d. Klein Koslau, Kreis Neidenburg) w świetle publikacji i materiałów archiwalnych // Wiadomości Archeologiczne. LXII. Warszawa, Ziemlińska-Odojowa W. Niedanowo. Ein Gräberfeld der Przeworsk- und Wielbark-Kultur in Nordmasowien // Monumenta Archaeologica Barbarica. VII. Kraków, Żórawska A. Wielokulturowe cmentarzysko w Starym Targu w świetle dotychczasowych publikacji, archiwaliów i ocalałych materiałów // Wiadomości Archeologiczne. LVII ( ). Warszawa, Алексеева Е. М. Античные бусы Северного Причерноморья / САИ. Г1 12. М., Алексеева Е. М. Античные бусы Северного Причерноморья / САИ. Г1 12. М., Фёдоров Г. Б. Два черняховских могильника в Молдавии // Материалы и исследования по археологии юго-запада СССР и Румынской Народной Республики. Кишинёв, Гопкало О. В. Бусы и подвески черняховской культуры. Киев, 2008.

316 316 Radosław PROCHOWICZ Радослав ПРОХОВИЧ Сердоликовые бусы в вельбарском могильнике Шелкув-Новы (Польша) «сувенир из Причерноморья»? Резюме Раскопками могильника Шелкув-Новы (ок. 70 км к северу от Варшавы) начиная с 2006 г. открыто ок. 200 погребений с трупоположениями и трупосожжениями пшеворской и вельбарской культур (фаза B римского времени начало эпохи Великого переселения народов). Особого внимания заслуживают погребения 130 (рис. 1, 2) и 133 (рис. 3, 4), которые по фи булам относятся к периоду C3 D. Открыты многочисленные детали костюма, которые в то время редко попадали в погребения. Отметим бусы из стекла, сердолика и янтаря (рис. 1, 2). Интересны украшения из погребения 130. Каменные бусы, найденные между се ребряными фибулaми, могут быть остаткам ожерелья. Стеклянные и янтарные бусины, се ребряные кольца и лунницы из медного сплава это остатки или свободных нитей с украшениями, или декора, нашитого на ткань или кожy. Редкое явление три фибулы в одном погребении. Если умершего похоронили головой на север по вельбарскому обряду, то одна фибула находилась в районе плеч и две у бёдер. Фибулы этой группы (рис. 3. 1, 2) вероятно, незавершённые, никогда не использовавшиеся копии. Следы необычных ритуалов тёмные прослойки в верхней части заполнения объекта, содержащие небольшие обожжённые камни, расплавленные фраг менты и отдельные кальцинированные кос ти. Среди погребального инвентаря были вель барские и черняховские вещи, которые, в сочетании с необычными обрядами, могли указывать на неполную ассимиляцию умерших. 14-гранные бусы из полудрагоценных камней (сердолика?), относительно редкие в центрально-европейском Барбарикуме, происходят в основном из районов пшеворской и вельбарской культур (известно ок. 220 экземпляров на 39 памятниках) (рис. 5). В междуречье Одера и Буга они попали c юга и юговостока, из районов сарматского и черняховского населения, в период C римского времени и в начале эпохи Великого переселения народов. Это может указывать на перемещение некоторых групп населения из Причерноморья. Возможно, 14-гранные каменные бусы из погребений 130 и 133 часть немногочисленных «сувениров с причерноморской родины».

317 317 Oleg RADYUSH The Second and Third Century Knob Spurs (Knopfsporen) in the Middle and Upper Dnieper Area Horse trappings and rider s equipment are important components of the ancient peoples weapons system. The most popular category of the equipment in the middle and upper Dnieper area throughout the period under analysis was spurs. Spurs probably appeared in the middle Dnieper in the first century BC and early first century AD when they occurred in the sites of the Zarubintsy culture (Obolon, Monastyryok) and in the cemetery of Mutin (Радюш 2011; Terpilovskiy, Zharov 2012). All the early finds belong to type Knopfsporen, widely distributed in the monuments of the late La Tène and pre-roman periods in Central Europe (Jahn 1921: 5 17; Luckiewicz 2006: ; Bochnack 2005: ; Филиповиħ 2011: ). Jerzy Ginalski has studied their further development from the first to the third century AD against the background of the Przeworsk culture (Ginalski 1991). The distribution of spurs in Eastern Europe is ill-studied due to a small number of sources in possession. Before the Second World War, the catalogues of the Kiev Historical Museum kept records of eight bronze spurs from pre-revolutionary collections, but five of them disappeared during the war, so their construction remains unknown, one artefact with enamelled insets on knobs survived (fig ) (Спицын 1903: 176, рис. 234), and two objects more had non-enamelled knobs. Two of these spurs were probably discovered in Kanev uyezd (pre-revolutionary administrative division) of the Kiev governorate, and one spur, undocumented and without enamelling, was a part of Faddey Kunderevich s collection (Корзухина 1978: 50). In post-war period, the number of finds increased but not significantly. Spurs from the Oka river area became known: 1956 investigation of a cemetery near village Razhki in the Penza oblast (administrative area) discovered a bronze spur in burial 6 (Полесских 1959: рис ) excavations of multi-layered settlement from the Roman and Great Migration periods near the village of Novyye Bezradichi in Obukhov district of the Kiev oblast discovered an iron knobbed spur (Даниленко 1976: рис. 9. 2). Simultaneously, two bronze spurs with enamel insets were unearthed in burial 28 of the cemetery of Skalistoye III in the Crimea (Богданова, Гущина, Лобода 1976: рис , 49). The survey of the late Zarubintsy settlement of Bobritsa in the Kiev-Svyatoshin district of the Kiev oblast discovered one bronze spur (Максимов 1991: рис. 9 10). These finds compose almost all the sources on the spurs of type Knopfsporen. Besides, there also are spurs with hooked ends, or Hakensporen, found at the monuments of the late Zarubintsy and early Kiev cultures (Казанский 1997; Радюш 2006), but in most cases this type has a wider chronology. Very small number of spurs in general and of the type under study in particular discovered in the Dnieper area is well understandable from Valeriy Perkhavko s paper presenting the only in Soviet scholarship analysis of Eastern European spurs in the first millennium AD (Перхавко 1978). In 1978, he knew only ten first fifth century spurs in the entire middle and upper Dnieper area, including only one knobbed spur from Novyye Bezradichi (Перхавко 1978: рис. 1). All the others are spurs from the Przeworsk burials in Western Ukraine, hooked spurs, and spurs of the later, Chernyakhov period. Michel Kazanski underlined the same situation in the late 1990s, when new spurs were not recorded in his list of the weapons in the Kiev culture (Казанский 1997: ). Although there are publications making the standing of the finds from the Chernyakhov period more or less clear (Kazanski, 1994; Магомедов, Левада 1996; Петраускас, 1999), pre-chernyakhov spurs, related to the so-called late Zarubintsy and early Kiev culture sites and monuments, still remain an ill-

318 318 Oleg RADYUSH studied phenomenon. Few finds do not allow colleagues to understand their variety and distribution. In 2006, I made an attempt to clarify the eastern border of the distribution of spurs in the Roman period, as it appeared from the finds known by the moment, and mentioned new materials related to riders equipment from the second to the fourth century which appeared in the Seym river, Don, and upper Oka areas (Радюш 2006). From the mid-2000s, the number of sources has been snowballing, but, to the great regret, the overwhelming majority of them are not from professional excavations or surveys. At the same time, during last five years I have recorded a great part of the finds determining their localization or region of origin. 15 finds were obtained by Belorussian and Russian museums. So far I know 30 spurs with inset enamel decorations (including one find from Lithuania) and more than 160 knobbed spurs without enamelling of the types distributed in the second and third centuries in the Dnieper area under analysis. The selection in possession allows me to analyse typological aspects of this equipment, but the lack of data regarding archaeological context of most finds does not allow ethnic, cultural, and chronological interpretation of them. The given paper deals with 69 specimens, first of all published and stored in museums, all enamelled spurs and those without enamelling that show various variants of design, with as precise localization as possible when it is possible. No doubts, the gaps in our knowledge will be refilled and clarified in the future. 1 The data in possession make it possible to map the finds to determine distribution areas of the second and third century knobbed spurs in Eastern Europe. They are related primarily with the forest and forest-steppe areas on the Dnieper and in the Baltic Sea area (fig. 1). The most saturated with finds are the Middle Dnieper area (Kiev and Chernigov oblast s, Ukraine) and the left bank of the Dnieper (Kursk and Belgorod oblast s, Russia, Sumy, Chernigov, and Kharkov oblast s, Ukraine). Although the finds are rare in the Upper Dnieper and the Desna areas, they do not occur there by accident, as it comes from the enamel circle treasures. The spurs are found in the Vitebsk, Grodno, and Gomel oblast s of Belorussia, Smolensk, Kaluga and Bryansk oblast s of Russia. The eastern border of the main distribution zone of knobbed spurs is in the upper reaches of the Psyol, Seym, and Severskiy Donets rivers. However, there are two finds indicating much wider zone of their distribution: the easternmost find comes from the cemetery of Rozhki located on the upper Moksha river, right tributary of the Oka, in the Penza oblast, and the northernmost one from Bol shoye Davydovskoye cemetery in Gavrilov Posad district of the Ivanovo oblast of Russia, located near the source of the Klyazma river, left tributary of the Oka. Although we know a pair of matching spurs from the Crimea, namely a grave in Skalistoye III cemetery, it is probably an exception, as knobbed spurs are generally not known in the steppe south and east of the Dnieper and in Caucasus (fig. 1). 2 The most complete corpus of artefacts with champlevé enamelling reflecting the situation in the early 1970s singles out three regions of their concentration: middle Dnieper area, eastern Baltic Sea area, and Masurian Lakeland (Корзухина 1978: inset map; 84 85). After that publication, the growing number of finds makes wider distribution of the artefacts of interest obvious. Apart from the northern Black Sea area, forest-steppe area on the left bank of the Dnieper stands out on the southeast of the whole area (Обломский, Терпиловский 2007). There are Usukh hoard and isolated finds in the Desna and Bryansk areas (Шинаков 2008; Ахмедов, Обломский, Радюш 2013), as well as on the upper Desna and Dnieper in the Smolensk oblast and Belorussia (Харитонович, Мяделец, Кенько 2008). Isolated finds and hoards of artefacts in this style are known on the east: in the middle Don and Khopyor areas (Обломский, Терпиловский 2007; Акимов 2012; Зиньковская 2011) and as far as the middle Volga and Kama areas (Бугров 1994), though individual artefacts occur 1 Although we should naturally take into account the particulars of the materials in possession, which is presented to a great extent by the finds with wide localization, it generally introduces a little change into the distribution areas discovered. 2 We should only note the only fragment of an iron spur, most likely of type Leuna or Szwajcaria, discovered in the destruction layer of Tanais from the mid-third century, which also contained several weapons of Germanic appearance. In the east of the forest-steppe zone, on the frontier with the steppe, there also are several finds of hooked spurs from the settlements with ceramic ware of the Kiev culture tradition from the third and fourth century, located on the Severskiy Donets and Khopyor rivers (Podgornoye, Shilovskiy, Yezdochnoye, Zanki, Raznobrychka) (Радюш 2006).

319 Fig. 1. Distribution of spurs, drinking horns, and whips in Eastern Europe. Spurs: 1 2 Skalistoye, burial 28; 3 4 Kanev uyezd; 5 Bobritsy; 6 8 middle Dnieper area (Kiev district?); 9 Novyye Bezradichi; 10 Kovray; Ukraine; 15 Priluki; Chernigov oblast ; Sumy oblast ; 22 Belopol ye; Kursk or Belgorod oblast ; 25 Shebekino district; Kharkov oblast ; 31 Pyatnitskoye; Belgorod oblast ; 34 Gochevo; 35 Kartamyshevo; Psyol river, the border between Kursk and Belgorod oblast s; Komarovka; 41 Korenskoye; 42 Chapli; 43 Seym river, Kursk district; 44 Bryansk oblast ; 45 Verdikhovo; 46 Bol shoye Davydovskoye, burial 3; 47 Razhki, burial 6; 48 border between Kaluga and Smolensk oblast s; 49 Lithuania; 50 Ostrino; 51 Ostrovets district, Grodno oblast ; 52 Rudevichi, Grodno oblast ; 53 Belorussia; 54 Mogilyov oblast ; 55 Vorotynsk; 56 Shishki; 57 Popovo-Lezhachi; 58 Vetka district. Drinking horns: A Moskalyovka; B Kursk oblast ; С Lezgur; D Usukh; E Moshchiny; F Chashniki district; G Linkaičių; H Maudžiorų; I Liel-Pudeŗi, J Kursk district; K Žadavainių; L Zaozyor ye / Lapsau; М Salinieki; N Kentes Kalns; O Bel sk fortified settlement. Whips: D Usukh; E Moshchiny; F Chashniki Signs on the map: Spurs: a variants Ia, Ib; b variants Iae, Ibe; c variants IIa, IIb; d variants IIae, IIbe; e variants IIIa, IIIIb, IIIc; f variants IIIae, IIIbe, III ce; g whips; h drinking horns; i main distribution areas for the circle of enamels; j localization as district; k localization as oblast or wider

320 320 Oleg RADYUSH

321 The Second and Third Century Knob Spurs (Knopfsporen) in the Middle and Upper Dnieper Area Fig. 2. Variants of the second and third centuries AD knobbed spurs in Eastern Europe: 1 Vorotynsk; 2 Bryansk oblast ; 3 Korenskoy; 4 5 Middle Dnieper area, Ukraine; 6 Vinnitsa oblast ; 7 Belgorod oblast ; 8 Ivnya district; 9 Shebekino district; 10 Kursk oblast ; 11 Belgorod oblast ; 12 Kursk oblast ; 13 Ukraine; 14 Priluki; 15 Belorussia; 16 Mogilyov oblast ; 17 Ostrino; 18 Razhki, burial 6; 19 river Psyol; 20 Bryansk oblast ; 21 Kiev oblast ; Ukraine; 24 Russia; 25 border between Kaluga and Smolensk oblast s; 26 Verdikhovo; 27 Sumy oblast ; 28 Lithuania; 29 Pyatnitskoye; 30 Chernigov oblast ; 31 Kharkov oblast ; 32 Kursk oblast ; 33 Psyol river; 34 Kharkov oblast ; 35 Psyol river; 36 Sumy oblast ; 37 Ukraine; 38 Sumy oblast ; 39 Russia, 40 Psyol river; 41 Chapli; 42 Komarovka; 43 Ostrovets; 44 Ukraine; 45 Sumy oblast ; 46 Ukraine; 47 Belopol ye; 48 eastern Ukraine; 49 middle Dnieper area; 50 Chernigov oblast ; 51 Kanev uyezd ; 52 Kharkov oblast ; 53 Gochevo; 54 Rudevichi 321 in the northern Black Sea area, Crimea, and North Caucasus (Корзухина 1978: ionset map; Bitner- Wróblewska 2011: fig. 1; Щукин 2002: 7). The merging of distribution maps of spurs and Eastern European enamels uncovers that spurs and enamels occur in the same areas on the Middle Dnieper and on the Left Bank of that river, as well as in the upper Dnieper and Desna regions. The finds of the second and third century spurs are well known in Lithuania, Masurian Lakeland in Poland, and Kaliningrad oblast of Russia, including those uncovered in assemblages and burials together with enamels (Michelbertas 2000; Волкайте- Куликаускене 1967; Vengalis 2005: pav ). In the first centuries AD, knobbed spurs demonstrated large variations of design, details, size, and technology. Particular elements were used in different combinations, with or without enamelling. The most important morphological differences of the spurs construction allow me to determine three basic variants indicated by Latin numerals (I, II, III). Sub-variants are marked by Latin symbols (a, b, c). Additionally, spurs with enamel insets are indexed as е. This preliminary division certainly requires extra work against wider selection of finds. 3 The first variant incorporates spurs with simple yokes, the second with platforms and transverse crests on yokes, the third with extensions and rosettes on yokes. Variant I. Spurs with simple yoke without extra elements (fig. 1. 1; 2. 5, 8). Knobs are circular or diamond-shaped, in one case star-shaped (fig ). Most spurs are solid bronze, a few pieces are composite with iron shank and bronze body. Pyramidal or conical shank with tetrahedral or circular crosssection is cm high, with different ornamentation: without enamelling or with enamel insets (fig ; 2. 5, 8). Ia. Yokes are elongated, measuring 6 8 cm between the knobs, with flat triangular cross-section, widening in the middle near the shank (fig ). The shank is short, up to 2 cm high, with circular or tetrahedral cross-section. Ib. Yokes are short (4 5 cm), shanks are solid and short. Knobs are flat and circular, of small diameter measuring up to 1 cm. 3 Unfortunately, the common problem for any research of enamelled finds is the dominance of found by chance artefacts, so their chronology is determined basically by the analogies in the Barbaricum. Only new researches and studies of the finds in close assemblages would allow us to clarify and particularize the chronology of knobbed spurs and to understand their origin and development.

322 322 Oleg RADYUSH Variant I also includes a fragment of yoke with star-shaped knob (fig. 1. 7). Iae. The construction is the same as of spurs without enamelling. Short yokes, short pyramidal shank with quadrangular cross-section, 2 3 cm high. Flattened circular knobs measuring up to 1 cm in diameter. Red enamel insets on platforms of knobs. Ibe. Short yokes, elongated pyramidal shank with quadrangular cross-section, 3.5 cm high. The only known specimen have diamond knobs and through holes in low shank (fig ). Red enamel insets decorate knobs, triangular vertical insets are on the shank (at least on two sides of it). Spurs of variant I are distributed in the upper Oka area (Kaluga oblast ), upper Dnieper area (Kursk oblast of Russia, Kharkov oblast of the Ukraine), and middle Dnieper area (Kiev, Chernigov, and northern Cherkassy oblast s of the Ukraine), one more find is registered in Lithuania. Analogies to the first variant of spurs are well known in Central Europe and eastern Baltic sea area (fig , 10 14, 16). They are most close to some variants of Ginalski s types E and F (Ginalski 1991: ryc. 11). The Vorotynsk spur is similar to type F1 from phase С1а, the form of the Popovo-Lezhachi spur is closer to type F3 from phases С1а С1b (Ginalski 1991: 64). Nonenamelled spur from Korenskoye (fig. 2. 3) and enamelled artefacts from the Chernigov oblast (fig ) have similar construction, which probably indicates their synchronism. Flat cross-section of yokes and complex enamelling on the shank, similar to some forms of variant IIIae, possibly indicate that these spurs are younger than those with solid yoke with triangular cross-section. Spurs of variant Ib are similar to Ginalski s type Е2, which date from the developed phase B2 to the phase C1a (Ginalski 1991: 62). There is only one bi-metallic spur, probably from Vinnitsa region, with the closest analogy in the Roman camp at Iža near Komárno in Pannonia Superior, or modern Slovakia (fig ) (Ardeleanu 2009: fig.7. 7). The spur from Belgorod oblast with star-shaped knob finds direct analogy in the former East Prussia (Tischler, Kemke 1902: Taf. XVI. 5). Variant II. Spurs with short yoke and rectangular platforms or flat transverse crests on both sides of yoke at the base of the shank. Most spurs are solid bronze, one item is composite with iron shank (fig ). Shanks with circular, quadrangular, or octahedral cross-section are conical, pyramidal, or fang-shaped in six cases (fig. 2. 9, 11, 13, 31, 32, 34). Knobs are mostly circular, but of three specimens have diamond and one circular knobs with four circular extensions (fig ). The size of knob greatly varies, from less than 1 cm in diameter to about 1.5 cm. There are both non-enamelled artefacts and those with enamelled insets on knobs and platforms, enamel is always red. IIа. With reduced crests or rectangular platform with triangular cuts along the edges on both sides of the shank base. Some pieces have small reduced crests on yokes (fig. 2. 9, 10, 37), though in most cases these crests merged. Two specimens (fig , 14) have specific figural triangular cuts on edges of the platforms. The finds of this variant include gigantic specimens with the shank 6 7 cm high (fig. 1. 2) and miniature artefacts with the shank 1.5 cm high (fig ). One bimetallic spur from the Kursk oblast has solid iron octahedral shank (fig ). Most spurs of the given variant are recorded in the upper Seym river and Psyol river areas (Kursk and Belgorod oblast s of Russia), one spur with cuts (fig ) originates from the Chernigov oblast, Ukraine. IIae. The construction is the same as variant IIa. Usually the platform is rectangular with triangular cuts, or in one case without cuts (fig ). Most cross-bars are reduced, in one case pronounced (fig ). The decoration consists of red enamelled insets on knobs (circular and in one case diamond) (fig ), triangular red enamel insets on the platform near the shank base (fig , 32). Discovered in the upper Seym and upper Psyol areas (Kursk and Belgorod oblast s of Russia) and in the Kharkov oblast (Ukraine). IIb. There are cross-bars, merged with subtriangle extensions to form a frame, located at the base of the shank (fig. 1. 3). All the finds are solid, bronze, with conical and pyramidal shanks with circular or quadrangular cross-section, 2 3 m high. All the finds originate from the Vitebsk and Grodno oblast s of Belorussia, the interfluve of the Viliya / Wilia and Berezina, and the Mogilyov oblast. IIbe. The construction is the same as variant IIb. Decorated with sockets with red enamelling on circular knobs. Cross-bars are degraded in one case, with holes on terminals (fig ). All the finds are recorded in Russia, in the upper Seym and Psyol areas. Generally, the area of distribution of spurs of variant II includes the upper Seym and Psyol (Kursk and Belgorod oblast s of Russia), Chernigov and Kharkov oblast s of the Ukraine, interfluve of the Vilia and the Berezina, Grodno and Mogilyov

323 The Second and Third Century Knob Spurs (Knopfsporen) in the Middle and Upper Dnieper Area 323 Fig. 3. Analogies and prototypes of the spurs in Central Europe and the Baltic Sea area. 1 Klein Tauersee; 2 3 Odry, barrow 22; 4 Macharen; 5 Sacherau; 6 Gronowo, barrow 6; 7 Loetzen; 8 Moythienen; 9 Baumgarten; 10 Katzelsdorf; 11 Drosig; 12 Tulln; 13 Sczwajzarija, barrow III, burial 1; 14 Samland; 15 Lauth; 16 Iza; 17 Vimose; 18 Kurisches Nehrung; 19 Klein Puppen; 20 Ramotten, burial 35; 21 Obliwiss; 22 Semeniskiu

324 324 Oleg RADYUSH oblast s of Belorussia, the only find is known in the Crimea. Only this variant includes a series of spurs with shanks up to 6 7 cm high, which makes it visibly different from other finds (fig , 13, 14, 31, 32). All of them originate from the left bank of the Dnieper (Chernigov and Kharkov oblast s of the Ukraine) and the Seym area (Kursk oblast of Russia). This group probably includes a bimetallic spur from the Kursk oblast (fig ): at any rate, the size of its shank base corresponds to the listed specimens. It is worth mentioning great similarity of the form of knobs of a spur (fig ) from Kursk oblast and the ending of the drinking horn with enamel insets (Радюш 2010: рис. 5. 5): it could indicate the same artisan s work. Kazimierz Godłowski has attributed platform spurs to group 2 of weapon graves in the Przeworsk culture from phase В2 of the Roman period (Godlowski 1992). Ginalski has singled out platform spurs from the Przeworsk materials as type E1 dating it back to the late phase В2 (Ginalski 1991: 61). A group of non-enamelled finds indicates a wide range of parallels to these spurs in the European Barbaricum (fig. 3. 9, 15, 18 22). Similar forms appear in Sambia (fig ), Lithuania (fig ), Przeworsk culture area (Ginalski 1991: 59, 61), and on the middle Danube (Beljak, Kolnik 2006: obr ). Ján Beljak and Titus Kolnik have supposed graves with spurs of similar forms to appear in the middle Danube area during the Marcomannic wars in relation to the troops coming from the Przeworsk and Wielbark territories (Beljak, Kolnik 2006: 87 88). Almost analogical to a spur from Kursk area (fig. 3. 1) is the find from the pre-revolutionary period from Prussia (Rositten cemetery, near modern settlement of Rybachiy, on the Kursh spit / Kurishes Nehrung), having the only difference, circular knobs of the Baltic artefact (Nowakowski 1996: 95; Archives: РМ III ). In Eastern Europe, there is only one close assemblage containing spur, in Skalistoye III cemetery in the Crimea (only one of two spurs among the grave goods is of variant II). The chronology of undercut grave 28-Л is determined as phase С1b according to glass beaker Eggers 247, 4 similar to the find in 1829 burial in Himlingøje (Lund- Hansen 1987: 207, fig. 134), and plate brooch from the end of the first quarter of the third century AD (Богданова, Гущина, Лобода 1976; Kazanski 1994: 431; Храпунов, Масякин 1998: 142). Spurs of variant II provided better fastening to boots due to wider platform and crossbars at the shank. Spurs with vertical crests from Scandinavian Vimose (Jahn 1921: 65, Abb. 67), Bolshoye Isakovo / Lauth (Kreis Königsberg) in the Baltic Sea area, 5 and the middle Danube area (Peška, Tejral 2002: , tab. 50, 54) probably were earlier variants. 6 The observations in possession suggest that pronounced cross-bars on yokes are older feature though entire platforms or merged cross-bars are later. Variant III incorporates spurs with extensions and rosettes at the shank base and on yokes and spurs with extensions in floriate baroque style. IIIа. Symmetrical extensions at both sides of the shank, short (fig , 21 27) or long (fig ). Extension ends are either flat or with semicircular terminals, in one case with holes (fig ). Short conical and pyramidal shanks circular or quadrangular cross-section are 2 3 cm high. Knobs are flat circular, semi-globular circular, and diamond-shaped. This group includes three bi-metallic artefacts (fig , 22, 24). Such spurs are found in the Ukrainian middle Dnieper area, Desna area (Russian Bryansk oblast ), upper Seym, upper Psyol, and middle Oka areas (Penza oblast of Russia). IIIae. The construction is the same as variant IIIa. Extensions are in the form of two symmetrical rosettes on either side of the shank (fig , 45), or four rosettes around the shank (fig ). Shanks are short, with circular or quadrangular cross-section. Small rosettes are about 1 cm in di paper supplies an incorrect reconstruction of the form of the vessel. More detailed publication of the given assemblage is being prepared now by Vyacheslav Masyakin. 5 Burial 1СЛ This destroyed assemblage contained a sestertius of Marcus Aurelius, bronze fibula, buckle, and conical appliqués, so it dates back to phase B2b B2/C1a (the information and the drawing are kindly offered by the cemetery excavator Konstantin Skvortsov). 6 Throughout the early Roman period as long as phase B2/C1, Germanic Barbaricum jewellery widely applied vertical crests decorating ornaments and cloth details, weapons and military equipment. This style gradually disappeared from the area west of the Neman by the Late Roman period, though it possibly existed for longer in the eastern Baltic Sea coastl. These elements became wider by the late second century AD, i. e. in phase B2/C1a (Almgren 1923: Taf. II, IV, V, VI).

325 The Second and Third Century Knob Spurs (Knopfsporen) in the Middle and Upper Dnieper Area 325 ameter, flat circular or puck-shaped, similar to the knobs. Rosettes and knobs are decorated with sockets with red enamel, one spur is ornamented with sockets with white enamel and triangular insets with enamelling on the shank (fig ), Gochevo spur possesses visible triangular cavities for enamel on the shank. Two spurs with enamel knobs have small hemispherical extensions on either side of the shank (fig , 50), one of them is obviously bimetallic (fig ). These spurs occur in the upper Neman area (Grodno oblast of Belorussia), upper and middle Seym and Psyol areas (Sumy and Cherkassy oblast s of the Ukraine and Kursk oblast of Russia). IIIb. Extensions at the shank in baroque style in the form of floriate crosses (fig ). Transverse extension on the yoke. Short bi-conical shank, up to 1.5 cm high, with circular cross-section. Flat circular knob. The only find originates from the border of the Kaluga and Smolensk oblast s. IIIbе. Similar to variant IIIb. One specimen, probably with iron shank, with long yoke and roughly made floriate crosses on long extensions below the shank (fig ), originates from the Kharkov oblast. Another find of this variant is solid cast, with extensions and knobs in the form of pronounced floriate crosses terminating with rosettes filled with red enamel, and with sockets between the rosettes filled with the same enamel (fig ). There are symmetrical triangular sockets filled with white enamel on the yoke at the shank. Pyramidal shank is short, with octahedral cross-section. The only find of the type originates from the upper Neman area (Grodno oblast of Belorussia). IIIc. Short projections at the shank, ending with circular bulges. Short shank with quadrangular cross-section is about 1 cm high. Diamond knobs decorated with four extensions each, with similar extensions on yoke symmetrical on either side of the shank (fig ). All the spurs of this variant are solid bronze. The finds originate from the Desna river area (Gomel oblast of Belorussia), the upper Dnieper area (Smolensk oblast of Russia), and the Oka river area (Ivanovo oblast of Russia). IIIce. Short, symmetrical, forking off or merged extensions on either side of the shank (fig , 51), which is with circular or sub-diamond crosssection. Flat circular or small diamond knobs are decorated with red enamelling. Both known spurs are solid bronze. The finds originate from the middle Dnieper area. A spur from burial 28 in the Crimean cemetery of Skalistoye III dates back to phase С1b, similarly to the above-described spur of type IIa from the same assemblage. Burial 6 in the cemetery on the river Moksha is dated to the early third century (Гришаков 2005: 18). Radiocarbon method dates burial 3 in Bol shoye Davydovskoye cemetery on the south of the Ivanovo oblast, with the votive complex containing two melted down remains of spur of variant IIIc, from the second half of the third to the early fifth century; taking the age of the wood into account, it could be 30 or 40 years younger, i. e. from the late third century to ca. 300 (Макаров, Красникова, Зайцева 2010: 49). Among the spurs of variant III there are finds continuing the development of variant Ia with long yoke (fig ), but most of them in do not meet with analogies in the European Barbaricum in contrast to first two variants. Their genesis was most likely related with the combination of details of simple knobbed spurs of variant I and elements of spurs of type Leuna that originated in the early second century, though variants with circular knobs and symmetrical extensions at the shank are similar to variants IIIa, IIIae, IIIce dating from phase С2 (Jahn 1921: 84; Giesler 1978: 22). The most striking features of spurs Leuna occur on some artefacts as wide flat circular rosettes near knobs (fig ), high extensions with hemispherical bulges (fig ), and symmetrical curved protrusions at the shank (fig ). By all appearance, extra extensions provided better fixation to boots, similar to variant II. Spurs of type Leuna with three riveted fasteners or knobbed spurs Szwajcaria with extra hook below the shank became more reliable in that regard than two-knob spurs. However, the eastern European artefacts display only imitations of such attachment, most likely with purely decorative purpose. Variant III is the latest in comparison with variants I and II. The find from Razhki cemetery suggests that the earliest specimens are from the early third century (fig ). Artefacts with enamelled rosettes date back to the first half of the third century (fig ). Two different spurs of types IIae and IIIae discovered together show their synchronism at some stage. Bi-metallic finds are probably relatively late (fig , 22, 24, 50). Among the latest specimens there also was a spur with trapezoid cross-section of yoke, which became a typical feature in the late period, and small extensions with holes (fig ). The inner development of variant IIIae possibly went from simple to complicated design (fig ) with four enamel rosettes at the shank (fig ) and most complicated specimens

326 326 Oleg RADYUSH Fig. 4. Goods from graves with spurs in Eastern Europe. A Skalistoye III cemetery, burial 28-Л (Богданова, Лобода, Гущина 1976); B Razhki cemetery, burial 6 (Полесских 1959)

327 The Second and Third Century Knob Spurs (Knopfsporen) in the Middle and Upper Dnieper Area 327 Fig. 5. Weeps and their parts. 1 Usukh, Briansk oblast (photo: the author; drawing: Sergey Voronyatov); 2 Chashniki district, Vitebsk oblast

328 328 Oleg RADYUSH with enamel insets on the shank (fig , 53). The latest spurs of variant III could be related to the degradation of enamel complex in general and the appearance of enamelled floriate forms (fig , 54); later on, there appeared non-enamelled forms with greatly degraded crests (fig ). There are specific spurs, three very similar finds, displaying an independent development but still keeping some features of variant III (fig ). The materials of Bol shoye Davydovskoye cemetery in Suzdal Fields, where one of those spurs was discovered, suggest their relation to the latest variations from the circle of enamels (Макаров, Красникова, Зайцева 2010: 50). The source of the formation of Eastern European variants of spurs obviously was Ginalski s types E F, widespread in Central Europe in phases B2 C1b, probably with participation of later constructions of spurs of type Leuna. Their later and independent development happened in the Dnieper and neighbouring areas. The spurs greatly varied similarly to those in Central European and Baltic cultures from the Roman period. Nevertheless, it is worth mentioning considerable difference of Eastern European variants from their western neighbours: as Ginalski has discovered against the background of wide selection of spurs from the Przeworsk sties, 7 the overwhelming majority of them are made of iron (Ginalski 1991). In the territory under analysis there exists an opposite picture: almost all the spurs are made of bronze alloys. It is worth noticing among particular features of the Eastern European tradition of horse rider s equipment wide use of composite constructions with the main body of bronze and shank of iron, with its greater part deeply sunken into the body. This detail is recorded in almost all the variants of spurs under study. A local specificity could explanation the great number of spurs with diamond knobs. In the Masurian Lakeland, they are discovered in grave 27 of the cemetery of Mojtyny / Moythienen (Hollack, Peiser 1904: Tab. V. b1, b2; Jahn 1921: Abb. 60). Besides, the finds from that cemetery included artefacts of the circle of enamels: brooches, pendants, and wreath (Hollack, Peiser 1904: 33, Abb ). However, few artefacts are known in other regions and out of contexts with enamels. Diamond-shaped knobs are also known in the Przeworsk and Wielbark sites (Ginalski 1991: ryc ; Peška, Tejral 2002: Abb , 2). The finds with and without enamel insets have a doubtless link in between: they both use various similar elements typical of the complex of Eastern European champlevé enamels in general: protrusions, cross-bars, geometric rosettes and knobs, cut-through ornamentation techniques, and others. The earliest knobbed spurs appeared in the Dnieper area in the period of the Marcomannic wars, i. e. the second half of the second century, though the main bulk of the finds belongs to the period from the late second to the mid-third century. The latest knobbed spurs of the variants under study probably existed in remote territories around the Dnieper, Desna, and Oka to the early fourth century, though other variants of horse rider s equipment related to spurs of types Leuna and Szwajcaria occurred in the main forest-steppe area on the Dnieper, which was occupied by the Chernyakhov culture then (Магомедов, Левада 1996: 310). First observations of the distribution of the Eastern European spurs in the second and third centuries has suggested that the horse-rider s equipment was widely distributed among the pre-chernyakhov population in the Dnieper area. The main population of this region consisted of the Kiev culture tribes and similar cultural groups (Обломский, Терпиловский 2007: рис. 3, 5; Лопатин, Фурасьев 2007: рис. 1). No doubts, south-eastern Baltic Sea area was closely related with the Dnieper region in the field of military innovation, as it comes from the analogous finds discovered in Sambia and on the left bank of the Dnieper. Deep ties are also uncovered by other elements of the enamel complex, like drinking horns, a feature of military culture of drinking feasts. In Eastern Europe, we know at list 15 finds of enamelled drinking horns (fig. 1). It is important to note a new category of horse rider s equipment and weapons recently discovered: bimetallic whips possibly related to the Roman tradition (flagrum). They occur in the south of the Vitebsk oblast, Belorussia, and in Usukh hoard, Bryansk oblast, Russia (fig. 5), on a bank of the Sev river, the left tributary of the Desna. Perhaps, such artefacts indicated the owner s high social status, being used both to drive the horse and in close combat. Although we still do not know the finds of horse harness from that period, future researches will possibly discover some elements of horse bridle among the complex of artefacts with enamelling. In 7 Ginalski s research lists 334 finds, including 303 from cemeteries, 14 from settlements, and 17 undocumented. Among them there are 307 iron artefacts (Ginalski 1991: 53).

329 The Second and Third Century Knob Spurs (Knopfsporen) in the Middle and Upper Dnieper Area my point of view, wide distribution of spurs, whips, and drinking horns in local antiquities is an impressive indication of the developed culture of warriors related to local horse-riding elite in the late second 329 and third centuries AD. It possibly appeared when the river road from the Baltic to the Black Sea, the most important one in Eastern Europe, began to be used actively. Catalogue of spurs found in Eastern Europe Belorussia 1. Vetka district (Gomel oblast ). Bronze. Variant IIIс. Stray find. Collection: not known. Not published (informer: Yevgeniy Vlasovets, Belorussia State University, Minsk). 2. Ostrino (Shchuchin district, Grodno oblast ) (fig ). Bronze. Variant IIb. Stray find. Collection: National History Museum of the Republic of Belorussia, Minsk (henceforth NHM). Not published. 3. Ostrovets district (Grodno oblast ) (fig ). Bronze, red enamel. Variant IIIae. Stray find. Collection: NHM. Not published (informer: Nikolay Plavinskiy). 4. Rudevichi (Zel va district, Grodno oblast ) (fig ). Bronze, red and white enamel. Variant IIIbe. Stray find. Collection: NHM. Not published (informer: Plavinskiy). 5. Mogilyov oblast (fig ). Bronze. Variant IIb. Stray find. Collection: not known. Not published (informer: Vlasovets). 6. Shishki (Slonim district, Grodno oblast ). Bronze. Variant IIIа. Stray find. Collection: not known. Not published (informer: Vlasovets). 7. Kamenets district (Brest oblast ). Bronze. Variant IIa. Stray find. Collection: not known. Not published (informer: Vlasovets), 8. Belorussia (fig ). Bronze. Variant IIc. Stray find. Collection: not known. Not published (informer: Vlasovets). Lithuania 9. Lithuania (fig ). Bronze, red enamel. Variant Iae. Stray find. Collection: not known. Not published (informer: Linas Talamnunas, Vilnius University). Russia 10. Vorotynsk (Babynino district, Kaluga oblast ) (fig. 2. 1). Bronze. Variant Ia. Stray find in the settlement area with layers from the Early Iron Age and Roman period. Collection: State Military History Museum and Nature Preserve of Kulikovo Field, Tula (henceforth KP). Not published. 11. Verdikhovo (Pochinok district, Smolensk oblast ) (fig ). Bronze. Variant IIIc. Stray find on the settlement from the Late Roman period. Collection: Kursk State Regional Museum of Archaeology (henceforth: KMA). Published: Радюш 2010: Border between Kaluga oblast and Smolensk oblast (fig ). Bronze. Variant IIIb. Stray find. Collection: not known. Not published. 13. Bryansk oblast (fig. 2. 2). Bronze. Variant Ia. Stray find. Collection: not known. Not published. 14. Bryansk oblast (fig ). Bronze. Variant IIIa. Stray find. Collection: KMA. Not published. 15. Gochevo (Belaya district, Kursk oblast ) (fig ). Bronze, enamel. Variant IIIae. Stray find. Collection: KMA photo archive. Published: Радюш 2010: рис Komarovka (Korenevo district, Kursk oblast ) (fig..2:42). Bronze, enamel. Variant IIbe. Discovered in vicinity of Komarovka fortified settlement, in the Roman period partly occupied by unfortified settlement. Collection: KMA. Published: Радюш 2010: рис Kartamysheva (Oboyan district, Kursk oblast ). Bronze. Variant IIa. Discovered in the settlement area from the Roman and Great Migration periods. Stray find. Collection: in 2007 was at private collector s in Kursk. Published: Радюш 2010: рис Komarovka (Korenevo district, Kursk oblast ). Bronze. Variant IIа. Stray find in the area of fortified settlement, partly occupied by unfortified settlement with materials from the Late Roman period. Collection: KMA. Published: Радюш 2010: рис Korenskoy (Ryl sk district, Kursk oblast ) (fig. 2. 3). Bronze. Variant Ia. Stray find in a settlement with the ceramic ware from the Late Roman period. Collection: in 2007 was at private collector s in L gov town. Published: Радюш 2010: рис Popovo-Lezhachi (Glushkovo district, Kursk oblast ). Bronze. Variant Iа. Excavated by Andrey Oblomskiy in a settlement layer form the early Kiev culture. Collection: KMA. Published: Обломский, Терпиловский 1998.

330 330 Oleg RADYUSH 21. Chapli (Kurchatov district, Kursk oblast ) (fig ). Bronze, red enamel. Variant IIbe. Stray find in a multilayered settlement from the Roman and Great Migration periods. Collection: Kurchatov City Regional Museum. Not published. 22. Psyol river, the border of Kursk and Belgorod oblast s (fig ). Bronze, red enamel. Variant IIbe. Stray find. Collection: KMA. Not published. 23. Psyol river, the border of Kursk and Belgorod oblast s (fig ). Bronze, red enamel. Variant IIae. Stray find. Collection: KMA photo archive. Published: Радюш 2010: рис Psyol river, the border of Kursk and Belgorod oblast s) (fig ). Bronze, red enamel. Variant IIae. Stray find. Collection: KMA photo archive. Published: Радюш 2010: рис Psyol river (Kursk oblast ) (fig ). Bronze. Variant IIIа. Stray find. Collection: KMA photo archive. Published: Радюш 2010: рис Kursk oblast (fig ). Bronze, red enamel. Variant IIae. Stray find. Collection: КMA photo archive Published: Радюш 2010: pис Seym river, Kursk district (Kursk oblast ). Bronze. Variant IIa. Stray find. Collection: in 2007 at private collector s in Kursk. Published: Радюш 2010: рис Kursk oblast (fig ). Bronze. Variant IIa. Stray find. Collection: КМА photo archive. Not published. 29. Kursk oblast (fig ). Bronze, iron. Variant IIa. Stray find. Collection: КМА. Not published. 30. Ivnya district (Belgorod oblast ) (fig. 2. 8). Bronze. Variant IIa. Stray find. Collection: КМА photo archive. Not published. 31. Shebekino district (Belgorod oblast ) (fig. 2. 9). Bronze. Variant IIa. Stray find. Collection: КМА. Not published. 32. Belgorod oblast (fig ). Bronze. Variant IIа. Stray find. Collection: КМА. Not published. 33. Belgorod oblast (fig. 2. 7). Bronze. Variant: not specified, possibly Ia. Stray find. Collection: КМА. Not published. 34. Razhki (Nizhniy Lomov district, Penza oblast ) (fig ). Bronze. Variant IIIa. Burial 6 in the cemetery excavated by Mikhail Polesskikh in Collection: Penza State Regional Museum. Published: Полесских 1959; Полесских Bol shoye Davydovskoye (Gavrilov Posad district, Ivanovo oblast ). Bronze. Variant IIIс (exact analogy to the Verdikhovo find). Two melted fragments of knob and yoke. Burial 3 in the cemetery excavated by Nikolay Makarov in Collection: Institute of Archaeology of the Russian academy of Sciences. Published: Макаров, Красникова, Зайцева: Russia (fig ). Bronze, iron. Variant IIbe. Stray find. Collection: not known. Not published. 37. Russia (fig ). Bronze, iron. Variant IIIa. Stray find. Collection: КМА photo archive. Not published. Ukraine 38. Bobritsa (Kiev-Svyatoshin district, Kiev oblast ). Bronze. Variant IIIa. Stray find in the late Zarubuntsy settlement. Collection: not known. Published: Максимов 1991: рис Vinnitsa oblast (fig. 1. 6). Bronze, iron. Variant Ib. Find spot not known. Collection: not known. Not published. 40. Novyye Bezradichi (Obukhov district, Kiev oblast ). Iron. Variant Ib. Settlement layer from the late Zarubintsy culture, excavated by Valentin Danilenko. Collection: not known. Published: Даниленко, 1976 рис Former Kanev uyezd, Kiev governorate. Bronze. Variant Ia. Stray find from the nineteenth century. Collection: National Museum of Ukrainian History (henceforth: MUH), Kiev. Published: Ханенко 1901: таб ; Корзухина 1978: Former Kanev uyezd, Kiev governorate (allegedly Perepel chitsy village) (fig ). Bronze, red enamel. Variant IIIсe. Stray find from the nineteenth century. Collection: MUH. Published: Корзухина 1978: 68, табл Kiev oblast (fig ). Bronze, iron. Variant IIIa. Stray find. Collection: КМА. Not published. 44. Middle Dnieper area, Ukraine (fig. 2. 4). Bronze. Variant Ib. Stray find. Collection: not known. Not published. 45. Middle Dnieper area, Ukraine (fig. 2. 5). Bronze. Variant Ib. Stray find. Collection: not known. Not published. 46. Middle Dnieper area, Ukraine (fig ). Bronze, red enamel. Stray find. Collection: not known. Not published. 47. Skalistoye III (Bakhchisaray district, Crimea). Bronze, red enamel. Variant IIae. Undercut grave 28-Л of the cemetery. Collection: Bakhchisaray Historical and Cultural Preserve (henceforth: BICP). Published: Богданова, Гущина, Лобода 1976: рис Skalistoye III (Bakhchisaray district, Crimea). Bronze, red enamel. Variant IIIae. Undercut grave 28-Л of the cemetery. Collection: BICP. Published: Богданова, Гущина, Лобода 1976: рис

331 The Second and Third Century Knob Spurs (Knopfsporen) in the Middle and Upper Dnieper Area Belopol ye, Sumy oblast (fig ). Bronze, red enamel. Variant IIIae. Stray find. Collection: not known. Not published (informer: Maxim Levada). 50. Sumy oblast (fig ). Bronze. Variant IIIa. Stray find. Collection: not known. Not published. 51. Sumy oblast (fig ). Bronze, red enamel. Variant IIae. Stray find. Collection: not known. Not published. 52. Sumy oblast (fig ). Bronze, red enamel. Variant IIae. Stray find. Collection: not known. Not published. 53. Sumy oblast (fig ). Bronze, red enamel. Variant IIIае. Stray find. Collection: not known. Not published. 54. Pyatnitskoye (Kharkov oblast ) (fig ). Bronze, red enamel. Variant Iае. Stray find. Collection: not known. Not published. 55. Kharkov oblast (fig ). Bronze, red enamel. Variant IIae. Stray find. Collection: not known. Not published. 56. Kharkov oblast (fig ). Bronze, red enamel. Variant IIae. Stray find. Collection: not known. Not published. 57. Kharkov oblast (fig ). Bronze, red enamel. Variant IIIbe. Stray find. Collection: not known. Not published. 58. Eastern Ukraine (fig ). Bronze, red enamel. Variant IIIae. Stray find. Collection: not known. Not published. 59. Kovray (Chernobay district, Cherkassy oblast ). Bronze, red enamel. Variant Iae. Stray find. Collection: archaeological room, gymnasium no. 116, Kharkov (when published). Published: Григорьянц, Дидык Priluki (Chernigov oblast ) (fig ). Bronze. Variant IIa. Stray find. Collection: not known. Not published (informer: Levada). 61. Chernigov oblast (fig ). Bronze, red enamel. Variant IIIаe. Stray find. Collection: not known. Not published. 62. Chernigov oblast (fig ). Bronze, red enamel. Variant Ibe. Stray find. Collection: not known. Not published. 63. Ukraine. Bronze. Variant Ib. Stray find, Faddey Kunderovich s pre-revolutionary collection. Collection: MUH. Published: Корзухина 1978: Ukraine (fig ). Bronze. Variant IIa. Stray find. Collection: not known. Not published. 65. Ukraine (fig ). Bronze, red and white enamel. Variant IIIae. Stray find. Collection: not known. Not published. 66. Ukraine (fig ). Bronze, red enamel. Variant IIIae. Stray find. Collection: not known. Not published. 67. Ukraine (fig ). Bronze, red enamel. Variant IIae. Stray find. Collection: not known. Not published. 68. Ukraine (fig ). Bronze, iron. Variant IIIa. Stray find. Collection: not known. Not published. 69. Ukraine (fig ). Bronze. Variant IIIa. Stray find. Collection: not known. Not published. Bibliography Almgren O. Studien uber nordeuropaische Fibelformen der ersten nachristlische Jahrhunderts mit Berucksichtigung der provinzialromischen und sudrussischen Formen // Mannus-bibliothek. 23. Leipzig, Archives of Herbert Jankuhn and Rudolf Grenz at the Regional Museum of Archaeology at Gottorf Palace, Schleswig, Germany. Ardeleanu M. Pintenii germanici din barbaricum (sec. II III p. chr.) // Crisia. XXXIX. Oradea, Beljak J., Kolnik. T. Germánske hroby z c aky a gbeliec. Príspevok k osídleniu juhozápadného Slovenska v dobe rímskej // Slovenská archeológia. LIV. 1. Bratislava, Bitner-Wróblewska A. East European Enamelled Ornaments and the Character of Contacts between the Baltic Sea and the Black Sea // Inter Ambo Maria: Contacts between Scandinavia and the Crimea in the Roman Period. Kristiansand; Simferopol, Bochnak T. Uzbrojenie ludności kultury przeworskiej w młodszym okresie przedrzymskim. Rzeszów, Giesler U. Jungerkaiserzeitlichc Nietknopfsporen mit Dreipunkthalterung vom Typ Leuna // Saalburg Jahrbucher. XXXV. Saalburg, Ginalski J. Ostrogi kablakowe kultury Przeworskiej. Klasyfikacja typologiczna // Przglad Archeologiczny. 38. Wroclaw, 1991.

332 332 Oleg RADYUSH Godłowski K. Zmiany w uzbrojeniu ludnoŝci kultury przeworskiej w okresie wpływόw rzymskich // Arma et ollae. Łόdz, Hollack E., Peiser F. Das Gräberfeld von Moythienen. Königsberg, Jahn M. Der Reiternsporn. Seine Entstehung // Mannus Bibliothek. 21. Leipzig, Kazanski M. Les éperons, les umbo, les manipules de boucliers et les haches de l époque romaine tardive dans la région pontique: origine et diffusion // Beiträge zur römischer und barbarischer Bewaffnung in den ersten vier nachchristlichen Jahrhunderten. Lublin; Marburg, Luczkiewicz P. Uzbrojenie ludności ziem Polski w w młodszym okresie przedrymskim / Archaeologia militaria. II. Lublin, Lund Hansen U. Römischer import im Norden. Warenaustausch zwischen dem Römischen Reich und dem freien Germanien // Nordiske Fortidsminder. Ser. B. 10. København, Michelbertas M. Die bronzensporen der Römischen kaiserzeit in Litauen // Svperiores barbari: księga pamiątkowa ku czci Profesora Kazimierza Godłowskiego. Krakow, Nowakowski W. Das Samland in der römischen Kaiserzeit und seine Verbindungen mit dem römischen Reich und der barbarischen Welt // Veröffentlichung des Vorgeschichtlichen Seminars Marburg. 10. Marburg; Warszawa, Peška J., Tejral J. Das germanische Königsgrab von Mušov in Mähren Mainz, Terpilovskiy R., Zharov G. A Warrior Cemetery from the Turn of the Eras on the Seym River // Inter Ambo Maria. Northern Barbarians from Scandinavia towards the Black Sea (abstracts). Simferopol, Tishler O., Kemke H. Ostpreussische Altertumer aus der Zeit der grossen Graberfelder nach Christi Geburt. Konigsberg, Vengalis R. Semeniškių 2-oji neįtvirtinta gyvenvietė // Archeologiniai tyrinejimai lietuvoje 2008 metais. Vilnius, Акимов Д. В., Ененко Е. А. Случайные находки вещей римского времени в пределах лесостепного Подонья // Stratum plus. 4. Кишинёв, Ахмедов И. Р., Обломский А. М., Радюш О. А. Клад из Суземского района Брянской области //Археологические исследования в Еврорегионе «Днепр» в 2012 г. Гомель, Богданова Н. А., Гущина И. И., Лобода И. И. Могильник Скалистое III в Юго-Западном Крыму // СА Бугров Д. Г. Предметы с выемчатой эмалью из Нижнего Прикамья // Историко-археологическое изучение Поволжья. Йошкар-Ола, Волкайте-Куликаускене Р. Успехи археологической науки в Литовской ССР // СА Григорьянц М. Н., Дидык В. В. Два уникальных предмета круга варварских эмалей из Днепровского Левобережья // Харьковский археологический сборник Гришаков В. В. Население верховьев Мокши и Суры накануне средневековья. Саранск, Даниленко В. М. Пiзньозарубинецьки пам ятки киiвського типу // Археологiя. 19. Київ, Зиньковская И. В. О новом ареале украшений круга выемчатых эмалей // РА Казанский М. М. Оружие Киевской культуры // Памятники старины, концепции, открытия, версии. 2. СПб.; Псков, Корзухина Г. Ф. Предметы убора с выемчатыми эмалями V VI в. н. э. в Среднем Поднепровье / САИ. Е1 43. М., Лопатин Н. B, Фурасьев А. Г. Северные рубежи раннеславянского мира в III V вв. н. э. // РСМ. 8. М., Магомедов М. Е., Левада М. Е. Оружие черняховской культуры // МАИЭТ V. Макаров Н. А., Красникова А. М., Зайцева И. Е. Могильник Большое Давыдовское 2 погребальный памятник первой половины I тыс. н. э. в Суздальском Ополье // РА Максимов Е. В. Актуальнi питання вивчення зарубинецькоï культури // Археологiя Обломский А. М., Терпиловский Р. В. Поселение Попово-Лежачи-4 и его место среди памятников киевской культуры // Acta Universitatis Łodziensis. Folia archaeological. 22. Łodz, Обломский А. М., Терпиловский Р. В. Предметы убора с выемчатыми эмалями на территории лесостепной зоны Восточной Европы (дополнение сводов Г. Ф. Корзухиной, И. К. Фролова и Е. Л. Гороховского) // Памятники киевской культуры в лесостепной зоне России (III V вв. н. э.) / РСМ. 10. М., 2007.

333 The Second and Third Century Knob Spurs (Knopfsporen) in the Middle and Upper Dnieper Area 333 Перхавко В. Б. Появление и распространение шпор на территории Восточной Европы // СА Петраускас О. В. Шпоры из могильника черняховской культуры у села Великая Бугаевка // 100 лет черняховской культуре. Киев, Полесских М. Р. Ранние могильники древней мордвы в Пензенской области // СА Полесских М. Р. Ражкинский могильник. Археология восточно-европейской степи. Саратов Радюш О. А. Находки шпор римского времени к востоку от Днепра. // Археологическое изучение Центральной России. Липецк, Радюш О. А. Украшения и снаряжение круга восточноевропейских выемчатых эмалей в бассейне верхнего течения Псёла и Сейма // Лесная и лесостепная зоны Восточной Европы в эпохи римских влияний и Великого переселения народов. Конференция. 2/2. Тула, Радюш О. А. Предметы вооружения и кавалерийского снаряжения на памятниках зарубинецкой культуры // Военная археология. 2. М., Спицын А. С. Предметы с выемчатою эмалью // Записки Императорского русского археологического общества. Русское отделение. V. СПб., Филиповиħ В. Мамузе из млаħег гвозденог доба у Србиjи // Гласник српског археолошког друштва. 25 / Београд, Ханенко В. И. Древности Поднепровья. IV. Киев, Харитонович З., Мяделец А., Кенько П. Новые находки изделий с выемчатыми эмалями на территории Беларуси // Матэрыялы па археалогіі Беларусі. 15. Мінск, Храпунов И. Н., Масякин В. В. Могила с двумя подбоями III в. н. э. из некрополя Дружное в Крыму // РА Шинаков Е. А. Находки предметов с эмалями позднеримской эпохи в Подесенье и вопросы их происхождения и атрибуции // Случайные находки: хронология, атрибуция, историко-культурный контекст (материалы конференции). CПб., Щукин М. Б. Об одной интересной находке из могильника на склоне Чатыр-Дага // Алушта и Алуштинский регион с древнейших времен до наших дней. Киев, Олег РАДЮШ Шпоры с кнопками (Knopfsporen) II III вв. в Среднем и Верхнем Поднепровье Резюме В статье рассмотрены многочисленные находки шпор с зацепами и кнопками II III вв. н. э., происходящие с территории Поднепровья и сопредельных регионов Восточной Европы. Эти элементы кавалерийского снаряжения позволяют характеризовать местные воинские традиции, сведения о которых до последнего времени оставались обрывочными. Появление шпор c кнопками ранних типов на территории Среднего Поднепровья может датироваться I в. до н. э. началом I в. н. э. на основании находок на зарубинецких па мятниках (Оболонь, Монастырёк) и в могиль нике Мутин. Находки следующего этапа связаны с позднезарубинецкими и киевскими памятниками. В приведённой сводке 69 экземпляров: цельнолитых бронзовых, бронзовых с декоративными вставками эмалей, а также ком по зитных биметаллических, с шипом из же ле за. Большинство находок случайные, как и значительное количество вещей горизонта «вос точноевропейских выемчатых эмалей», к которому они в целом относятся. Находки шпор с кнопками известны от верховий Днепра и Немана до Среднего Поднепровья и в отдельных случаях Поочья, на территории Литвы, Гродненской и Могилёвской об ластей Белоруссии, Смоленской, Брянской, Калужской, Ивановской, Пензенской, Курской и Белгородской областей России, Киевской, Черкасской, Черниговской, Сумской и Харь ковской областей Украины; две находки сделаны

334 334 Oleg RADYUSH в Крыму. Наибольшее скопление отмечено в Сред нем Поднепровье и на левобережье Днепра. Шпоры с кнопками II III вв. н. э. де монстрируют значительное разнообразие оформления, деталей, размеров и технологии изготов ления. Отдельные элементы ис поль зуются в различных комбинациях как в эмалевом, так в безэмалевом исполнении. Наиболее существенные морфологические различия в конструкции шпор позволяют предварительно вы делить три их основных варианта и ряд подвариантов. К первому варианту отнесены шпоры с простыми дужками, ко второму шпоры с площадками и поперечными гребнями на дужках, к третьему шпоры с выступами и розетками на дужках. Наиболее ранними являются на ходки I и II варианта, прототипы которых в Европе относятся к фазе B2. Источниками для формирования восточноевропейских вариантов шпор послужили широко распространённые в Центральной Европе типы (Ginalski E F), датирующиеся фазами B2 C1b. Вероятно, своё влияние оказали и более поздние конструкции шпор типа «Лёйна». Дальнейшее и са мостоятельное развитие они получили на территории Поднепровья и в соседних регионах. Особенностями восточноевропейской тради ции снаряжения является преобладание брон зовых образцов над железными, что кардинально отличается от западных германских культур, широкое применение композитной би металлической технологии, а также значи тельное количество шпор с эмалями и шпор с ромбическими кнопками. В декоре шпор с эмалями и без них используются мно гочисленные одинаковые элементы, харак терные для вещевого комплекса восточноевропейских выемчатых эмалей в целом отростки, поперечные перекладины, розетки и кнопки геометрических форм, приёмы прорезной орнаментики и др. Наиболее поздние шпоры с кнопками рассмотренных вариантов, возможно, дожи вают на удалённых территориях в верхнем Поднепровье, Подесенье и Поочье до начала IV в. н. э. В это время на лесостепной территории Поднепровья, занятой черняховской куль турой, распространяются новые варианты кава лерийского снаряжения, связанные уже с разновидностями шпор типа «Лёйна» и «Швайцария». В области военных инноваций юго-восточная Прибалтика была тесно связана с Поднепровьем, на что указывают аналогичные находки на территории Самбии и на левобережье Днепра. Такие же глубокие связи демонстрируют и другие категории эмалевого комплекса, на пример, ритоны, относящиеся к дружинной пиршественной культуре. Отмечено появление ра нее неизвестных категорий кавалерийского снаряжения биметаллических плетей, по всей видимости, имевших римское происхождение. Широкое распространение шпор, использование нагаек, наличие в местных древностях рогов для питья достаточно явно указывают на сложившуюся в конце II III вв. н. э. дружинную культуру, связанную с местной воинской всаднической элитой. Её появление, возможно, связано с началом активного использования важнейшего для Восточной Европы речного пути от Балтики к Чёрному морю.

335 335 Stanislav SHABANOV Roman Import or Germanic Influence? On the Problem of Sarmatian-Germanic Contacts in the Crimea According to Glass Vessels Account The aspects of Germanic tribes appearance and settlement in the Crimea has been being studied by researchers since the nineteenth century. Initially they had used written sources accounts (Khrapunov 2011: 102), though the twentieth century witnessed an important progress in investigation of Germanic presence in the Crimea in the Late Roman period, when the scholars obtained the materials of archaeological investigations of funeral sites in Ay-Todor, Chyornaya River, Sovkhoz no. 10, Druzhnoye, and Neyzats. There are special studies discussing ornaments, ceramic ware, and weapons from the cemeteries in the foothill and mountainous areas in the Crimea. Scholarly tradition relates some of the glass vessels found there with the migration of tribes of Germanic circle, though others probably formed elements of Roman import. The first group possibly includes beakers with cut ovals of Hans Jürgen Eggers types 223 and 230, though the second group contains vessels decorated with incised lines and geometric patterns from the Late Roman period, discovered both in the Barbaricum and in the Roman imperial territory. Some categories of glass vessels, such as rhyta or drinking horns, were formerly not interpreted as indicators of Germanic presence. Later on, I will go deep into specific categories of glass vessels and the problem of their way to the Crimean foothill area. 1. Eggers type 223 In the Crimea, there is only one glass beaker Eggers 223, discovered in undercut grave no. 139 of the cemetery of Neyzats located in the middle of the foothill area, on the right side of the Zuya river, 1 km south of Barabanovo village, Belogorsk district. It measures 13.5 cm in height and 13.2 cm in rim diameter. This hemispherical beaker (fig. 1.1) with straight tip of the rim and flat bottom is made of thick semi-transparent glass of olive tint. Its body is ornamented with three horizontal lines of cut ovals. Undercut grave no. 139 contained two burials, with the latest (of a child) from the first half of the fourth century AD (Храпунов 2004: 304). Glass beaker of similar form and decoration (fig. 1. 2) originates from funeral assemblage no. 100 in the Chernyakhov culture cemetery of Velikaya Bugayovka in vicinity of Kiev. It is different from the Crimean find due to its smaller size (measuring 9 cm in height and 10.5 cm in rim diameter), out-turned lip, and colourless glass. It probably dates to the first half of the fourth century AD (Петраускас, Пастернак 2003: 68). The beaker from grave no. 139 of Neyzats cemetery is similar to the vessel discovered near Sigersted in Seeland island, Denmark (fig. 1. 3). Different interpretations of the latter find have been suggested. Eggers has attributed it to his type 223 from phases С2 (AD ) С3 (AD /75), though such vessels were probably also used later, in phase D1 (Eggers 1951: 180, Taf. 15, Karte 57). Günter Rau has attributed the artefacts form the Sigersted assemblage (wooden bucket with bronze appliqués, gold finger-rings and cup) to the first half or middle of the fourth century AD (Rau 1972: 129, 170). Eldrid Straume has related this vessel to group 1 singled out by her (Straume 1984: 38), from phases C3 D1, questioning their appearance and existence in phase C3 (Straume 1987: 120, taf ). Ulla Lund Hansen has dated this beaker to phases С1b (AD 200/10 250/60) С2 (AD 250/60 310/20) of her Scandinavian timeline (Lund Hansen 1991: 100, 411). Having analysed imported glass vessels discovered in Scandinavia, Gunnar Ekholm has ascribed the Sigersted beaker to his own group II B-1, supplying this group of small, thin-walled bowls and beakers with round, oval, or elongated facets and sometimes wheel-cut lines with analo-

336 336 Stanislav SHABANOV Fig. 1. Glass beakers Eggers 223: 1 Neyzats; 2 Velikaya Bugayovka; 3 Sigersted gies among the Roman imports in Karanis (Ekholm 1963: 33). 2. Eggers type 230 So far we know finds of five complete and three fragmented beakers of Eggers type 230 in the Crimean foothill area. One glass beaker (fig. 2; 3.1) (measuring 11.4 cm in height and 9.2 cm in rim diameter) with cylindrical body, flat bottom, and a little bit out-turned rim was discovered in burial vault no. 301 in the cemetery of Neyzats. The vessel is made of olive semi-transparent glass. The beaker s body has four horizontal rows of cut ovals and a line below the rim. The fourth century vault no. 301 contained six burials, but the vessel in question cannot be ascribed to one of them. Three more cylindrical glass beakers with four rows of cut ovals and two uneven lines below the rim originate from vault no. 275 in the cemetery of Neyzats. Made of olive glass, they were discovered crushed, though their form and shape we could reconstruct as: measuring 13.6 cm in height and 9.2 cm in rim diameter (fig. 3. 3); 15 cm in height and 10.2 cm in rim diameter (fig. 3. 4); 14 cm in height and 8.5 cm in rim diameter (fig. 3. 5). Burials were made into vault no. 275 throughout the fourth century AD (Khrapunov 2008: 214). More fragments of two thick-walled beakers of dark-yellow glass with cut ornamentation were discovered among the group of vessels outside graves no. 15 (fig. 3. 7) and in topsoil (fig. 3. 6) in the cemetery area (Храпунов, Мульд 2004: 309). Another beaker with cylindrical body (fig. 3. 2) and flat bottom was uncovered in vault no. 3 in the cemetery of Druzhnoye, 20 km south-east of Simferopol. The vessel has slightly out-turned rim, and is made of thick semi-transparent olive glass, with the body having four rows of cut ovals and one uneven line below the rim. It measures 9.7 cm in height and 6.7 cm in rim diameter. All the burials in vault no. 3 were made in the fourth century AD (Храпунов 2002: 16, 67). A fragment of thick-walled beaker (fig. 4) was unearthed in the cemetery of Almalyk-Dere at the foot of Mangup mountain in the south-western Crimea (fig. 2. 1). It originates from the fill of a ditch that contained backfills produced by plunderers and possibly remains of funeral feasts. In this fill there were fragments of hand-made and red-clay vessels and animals bones. The shards of amphorae Fig. 2. Glass beaker from Neyzats cemetery

337 Roman Import or Germanic Influence? On the Problem of Sarmatian-Germanic Contacts in the Crimea According to Glass Vessels Account 337 Fig. 3. Glass beakers Eggers 230: 1, 3 7 Neyzats; 2 Druzhnoye discovered there date back from the fourth to the seventh century (Shabanov 2011: 211). It is worth mentioning that early burial assemblages in this cemetery contain many finds from the final stage of the Chernyakhov culture (Mączyńska 2010: 91). 3. Straume s type VI; Ekholm s type Bremsnes This big beaker with cylindrical body and flat bottom of light-green glass was discovered in vault no. 78 in the cemetery of Druzhnoye (fig. 5. 1). The rim is slightly out-turned. On the body there are two incised uneven lines separating three ornamental rows, each filled with a line of cut ovals of different forms. Dimensions: 14 cm in height and 11 cm in rim diameter. Burials into vault no. 78 were made in the fourth century AD (Храпунов, Храпунов 1999: 252; Храпунов 2002: 33 34). Although I cannot suggest any precise analogy to this glass vessel, there are two beakers from in Norway, similar to the Crimean find by the way of separation of polished ovals rows by incised lines. They originate from assemblages in Gjerla and Bremsnes (fig ). Eggers has dated the Bremsnes beaker to phase D supposing its pro- Fig 4. Glass beaker from Almalyk-Dere cemetery

338 338 Stanislav SHABANOV Fig. 5. Glass beakers Straume VI: 1 Neyzats; 2 Gjerla; 3 Bremsnes duction in the Roman empire (Eggers 1951: 180, Kat. 323, Karte 58). Straume has attributed these two vessels to type VI dating them within periods С2 D and C3 D (Straume 1987: 77, 83, Taf , Karte 4. 2, 15). Ekholm has singled them out as a specific type called Bremsnes (Ekholm 1956: 51, Abb. 6k). Igor Gavritukhin has determined thickwalled truncated-conical beakers decorated with dense cut planes as type Lugi, separating them as series Bremsnes and agreeing with Straume s chronology (Гавритухин 1999: 52 53, pис , 4. 1, 19). 4. Straume s type VII of series A; Rau s type Gavrilovka 5 The excavations of Bakla Ravine cemetery in vicinity of modern Skalistoye village of Bakhchisaray district discovered a fragment of the lower body of glass beaker with polished ornamentation of dense ovals (fig. 6). It is made of green glass and probably had truncated conical body. Burials in this cemetery contained materials from the second half of the fourth to the ninth century (Айбабин, Юрочкин 1995: 125). Similar vessels are known in the Chernyakhov culture area from the late fourth to the fifth century AD (Сымонович 1957: 24, рис. 4. 8; Сымонович 1977: ; Гавритухин 1999: 49; Магомедов 2001: 65 66, рис ). Rau has included them to type Gavrilovka 5 dating from the last quarter of the fourth to the first quarter of the fifth century AD (Rau 1972: 166, Fig. 52). Straume has labelled analogous conical beaker from Högom in Sweden her type VII of series А from phase D2 (Straume 1987: 110). Gavritukhin has pointed out that the finds of such vessels are not known in Central Europe, concentrating mainly in modern Ukraine, Moldavia, and Romania, that is in the Chernyakhov culture area. Several beakers were discovered on the south of the Scandinavian peninsula (Gavritukhin 2011: 51). As Gavritukhin has pointed out, the finds of such vessels are not known in the pre-hunnic period (Гавритухин 2007: 18). Fig. 6. Fragment of glass beaker from Bakla Ravine cemetery

339 Roman Import or Germanic Influence? On the Problem of Sarmatian-Germanic Contacts in the Crimea According to Glass Vessels Account 339 Fig. 7. Glass beakers from Neyzats cemetery: 1 pit with vessels no. 11; 2 vault no Eggers type 195 One beaker (measuring 11.2 cm in height, 8.5 cm in rim diameter, and 2.4 cm in stem base diameter) (fig. 7. 1) was discovered in 2010 in pit with vessels no. 11 (Шабанов 2011: 153). This pit probably was a chambered construction, because there remained vertical slabs separating it on the west. Besides, all the vessels remain entire, though in pits they are usually crashed. Obviously, the ancients carved a niche into the wall of the pit to place vessels there. In total there were 10 hand-made, three red-slip, and one glass vessels (fig. 8). All the pits with vessels in the cemetery of Neyzats date back to the fourth century AD (Храпунов 2011: 22). Another vessel (13.5 cm in height, 11 cm in rim diameter, and 2.5 cm in stem base diameter) (fig. 7. 2) was uncovered in vault no. 485, containing burials from the fourth century AD. It was placed at the eastern wall of the vault chamber together with red-slip and hand-made vessels and crashed by fallen pieces of collapsed vault. In the vault there were four burials of adults total accompanied with many hand-made and red-slip vessels. Eggers has dated similar beaker with soldered threads from the cemetery of Salthammer in Norway to phase С2 (AD ) considering it to be a Roman import (Eggers 1951: 178, Taf ). The excavation of glass-making workshop in Cologne discovered the same stemmed tumbler with soldered ovals: Otto Doppelfeld has dated it to the third or fourth century (Doppelfeld 1966: fig. 132). 6. Rau s type Sigersted Ganzkow One glass vessel (fig. 10.1) was discovered in the cemetery of Suvorovo in vault no. 38, with burials made around the mid-fourth century AD (Юрочкин, Труфанов 2003: 218). The beaker (measuring 6.4 cm in height and 8 cm in neck diameter) has spherical body, round bottom, and outturned lip. The body is decorated with a single line of cut dashes and frieze of diamonds and ovals. Donald Harden has mentioned a find in Karanis in Egypt of a fragment ornamented like the Ganzkowo vessel (Harden 1936: pl. XVI. 428). 7. Rau s type Frunzovka One beaker (fig. 9, 10. 2) (measuring 10.2 cm in height and 9.7 cm in rim diameter) originates from vault no. 4 in the cemetery of Neyzats with the fourth century AD burials (Храпунов 2008: 385). Although I cannot find any exact analogy to this

340 340 Stanislav SHABANOV Fig. 8. Neyzats cemetery. Pit with vessels no. 11 glass vessel, the vessel with most close form and ornamentation originates from the Chernyakhov cemetery near modern village of Furmanovka (Odessa oblast, i. e. administrative area). Vladislav Kropotkin has dated it to the fourth century AD (Кропоткин 1970: 30, рис ). Rau has suggested more narrow date for the Frunzovka beaker, AD (Rau 1972: 167). One vessel with the same style of ornamentation is discovered in the cemetery of Kosanovo in grave no. 9 and attributed by Ol ga Gey and Igor Bazhan to chronological indicators of the third period of the Chernyalhov cultute (AD 310/20 350/55) (Гей, Бажан 1997: табл ). 8. Ising s form 113 In the Crimea we know only two glass drinking horns datable to the Roman period. One fragmented vessel (fig. 12.I 1) was discovered in undercut grave Л28 in the cemetery of Skalistoye III (Бог данова et al. 1976: рис. 11), with large bow one-piece fibula of the so-called Imkerman type with low string of Anatoliy Ambroz s group 15, series II. He dates these brooches to the second and third centuries AD (Амброз 1966: 52), though their maximum distribution was in the second quarter and middle of the third century AD (Храпу нов 2002: 52 54; Храпунов 2011: 27). The drinking horn is made of greenish glass, with the body decorated with thin glass threads of the same colour. Another vessel (fig. 12.I 2) was discovered in vault no. 7 in the cemetery of Sovkhoz 10 (or Sevastopol skiy, according to another version), where burials were made in the late fourth and fifth centuries AD. Analogous drinking horn with soldered glass thread (fig. 12.I 3) was discovered in the cem-

341 Roman Import or Germanic Influence? On the Problem of Sarmatian-Germanic Contacts in the Crimea According to Glass Vessels Account Fig. 9. Glass beaker from vault no. 4 in Neyzats cemetery etery of Kranj in Slovakia from the fourth century AD (Lazar 2003: , fig ). * * * There is only one fragment of thick-walled beaker (probably of Eggers type 230) from Greco- Roman sites in the Crimea: it was discovered by excavations of Pantikapaion in Nina Sorokina has dated it to the late third or fourth century AD (Сорокина 1962: 232). 1 Cylindrical glass beakers with four rows of cut ovals, of type Eggers 230, occur in many areas of eastern Europe, especially within the Chernyakhov 341 culture area (Кропоткин 1970: 30). When developing their single chronological system of the Gothic invasions period in cultures of Eastern Europe and Caucasus, Gey and Bazhan have attributed the appearance and distribution of these beakers to the third (AD 310/20 340/50) and fourth (mid-fourth century to AD 375) periods of the Chernyakhov culture (Гей, Бажан 1997: 48). Erast Symonovich has pointed out that the earliest specimens of Eggers 230 vessels in the Dnieper area have more accurate cutting, so he suggested their date as the second half of the third and fourth century AD (Сымонович 1977: 181). Emmanuil Rikman has mentioned that vessels with cut ovals of different forms and variations were the only or the predominant type of glass vessels among the barbarian population of the Dniester area (Рикман 1972: 95; Рикман 1975: 225, 227). In the analysis of glass vessels discovered in Pannonia, Lászlo Barkoczi has related the appearance of beakers of the type in Hungary with the late fourth and early fifth century AD (Barkoczi 1988: 105, taf. XV. 166). They are also known in the Eastern Danube area. On the north-east of Romania there were about 20 vessels with cut ovals allowing the researchers to suggest their import from production centres located in the Barbaricum rather than in the Roman imperial territory (Pánczál, Dobos 2007: 69, fig. 2). 2 There are different interpretations of the chronology of Eggers 230 beakers. This way, Eggers has studied the finds of these beakers in the area of free Germanic tribes to limit their existence to phases C1 (AD ) С2 (AD ), suggesting their existence in the fourth century as well (Eggers 1951: 180, Karte 58). Rau has dated beakers of type Kovalk to a later period, i. e. second and third quarters of the fourth century AD (Rau 1972: fig. 52). Galina Nikitina has accepted Rau s interpretation by placing beakers 1 In his big paper discussing the finds of cut glass beakers in the south of Eastern Europe Gavritukhin has supplied an image of one vessel of the type from Chersonesos (Gavritukhin 2011: fig. 4B. 3), but without mentioning it in the main body of the text. The same fragment has been published by Larisa Golofast among the finds of glass vessels from the early Byzantine layers of ancient Chersonesos (Голфаст 2001: рис ). It was discovered in assemblage from the middle or third quarter of the sixth century and attributed by the publisher to form 107 according to Clasina Isings classification (Голофаст 2001: 126, 127). 2 So far we know only one glass-making workshop outside the Roman empire. Excavations near modern village of Komarovo in the Chernovtsy oblast of the Ukraine discovered a settlement from the third and fourth century with sites of glass-making. There were multitude fragments of various vessels including thick-walled beakers with cut ovals. Researchers have no doubts that thin-walled transparent colourless glass vessels were not the only ones produced there (Смішко 1964: 75, 77; Щапова 1983: 153; Щукин 2005: 181), but still cut vessels were imported (Щапова 1978: , рис. 2.).

342 342 Stanislav SHABANOV Fig. 10. Glass vessels with cut strokes and geometric design in the Crimean foothill area: 1, 3 Suvorovo; 2 Neyzats; 4 Chyornaya River Fig. 11. Cut glass vessels excavated in the Roman glass-making workshop in Al ma-kermen fortified settlement

343 Roman Import or Germanic Influence? On the Problem of Sarmatian-Germanic Contacts in the Crimea According to Glass Vessels Account from the Chernyakhov cemetery Oselivka into the same chronological frames (Никитина 1995: 82). 3 Straume has suggested the period of Eggers 230 vessels in Scandinavia as phases С2 С3 (Straume 1987: 29, Taf. 2. 6, 35, 41, 46, 76), with their finds mainly concentrated in the south of Norway (Holland 2001: 37). Ekholm has made an analysis of imported glass vessels discovered in Norway to consider, since the most part of them originate in the south-eastern Europe, them to be of eastern origin (Ekholm 1963: 33). It is worth paying attention in this regard to Gavritukhin s interesting observations on a possible migration of some Chernyakhov population to Scandinavia about the late fourth or early fifth century, to escape from the Hunnic invasion. For example, he has used it as an explanation for the appearance of Straume VII beakers in Norway and Sweden (Gavritukhin 2011: 56 57). Therefore, we can be confident that glass Eggers 230 beakers appeared in the Crimea in the fourth century AD. All the finds of such vessels in the Crimean period date to that period. An earlier chronology is also possible (perhaps the second half or end of the third century AD), as it is suggested by the find (although the only one) of the fragment of glass beaker on the site of ancient Pantikapaion. Except for Neyzats cemetery, glass beakers with soldered ovals of type Eggers 195 are not known in synchronous barbarian sites or Roman period complexes in Greco-Roman cities and towns in the Crimea. Fragmented green glass bowl of similar form and ornamentation was discovered in a grave in the Chernyakhov cemetery of Krasnyy Mayak in southern Ukraine (Гошкевич 1913: 135, рис. 36). Kropotkin has dated this glass beaker to the fourth or first half of the fifth century AD (Кропоткин 1970: 30, рис ). This cemetery combines cremation and skeleton burials. The finds include redslip vessels and copper coin of emperor Theodosius I (AD ) (Сымонович, Кравченко 1983: 69 70) In Olbia, there are finds of shards of glass vessels with conical body and soldered threads of the same glass as the vessel itself dated to the third century AD by Sorokina (Сорокина 1976: 201, рис ). Glass beaker with a short flattened stem of brown glass with the walls decorated with relief ovals and vertical lines was discovered in grave no. 51 in the Chernyakhov cemetery of Kholmskoye in 343 southern Ukraine (Гудкова, Фокеев 1984: 70 71, рис. 21.1). The researchers have dated it to the fourth century AD with the earlier chronology still possible (Гудкова, Фокеев 1984: 83; Росохацкий 1987: 147). According to Symonovich, the glass vessels with soldered threads discovered in the Chernyakhov culture sites in southern Ukraine are from the third and fourth centuries (Сымонович 1977: ). One glass beaker with soldered ovals on its stem and glass drops on the walls was discovered in the cemetery of Kosino in western Ukraine, in a rich woman s burial (Кропоткин 1970: 103). It was accompanied with a silver tow-plated brooch with insets. Anatoliy Ambroz has attributed this fibula to group 21 of sub-group II variant Б dating it to the second half of the fifth century (Амброз 1966: 87, рис. 6. 2). Gavritukhin has conceded this brooch s chronology also in the last third of the fifth century (Гавритухин 1999: 59). Mark Shchukin has dated the assemblage from Kosino to the mid-fifth century (Щукин 2005: 253). Glass beaker similar to the Kosino find was uncovered from burial 20 in the cemetery of Independenţa in Romania. Its morphological similarity allows Gavritukhin to attribute it to the same type Kosino in his chronology of the final stage of the Chernyakhov culture (Гавритухин 1999: 57). Low beaker with uneven lip and three large ovals of soldered thread was discovered in burial VIII of the cemetery of Izvoare, Romania (Florescu, Daicoviciu, Roşu 1980: 200). It was accompanied with bone comb of Sigrid Thomas type II, from the fourth century (Thomas 1960: ), two plain two-plate brooches of Ambroz s group 21, subgroup I, variant АА, from the fourth century AD (Амброз 1966: 82). High conical beaker of light green thick glass, with truncated conical stem, ornamented with ovals of soldered threads was discovered in the grave assemblage of Salthammer, Norway. Eggers has dated it to phase С2 (AD ), attributing it to Roman imports (Eggers 1951: 178, taf ). Ekholm has also single out the Norwegian beaker to Greco-Roman imports with analogies in the territory of Hungary (Ekholm 1956: 54; Ekholm 1963: 34). Straume has dated the Salthammer assemblage to the end of phases С2 С3 (Straume 1984: 77; Straume 1987: 97). Lund Hansen has dated this beaker to phase С3 (AD 310/20 400) (Lund Hansen 1991: 442). 3 Nikitina has noted an important aspect: glass beakers with thicker walls and solid bottom are earlier than similar vessels with more transparent and thinner walls.

344 Fig. 12. I Glass drinking horns: 1 Skalistoye III; 2 Sovkhoz no. 10; 3 Kranj (Slovenia); 4 Shapka cemetery (Abkhazia). II Distribution Map of Germanic Glass Drinking Horns: 1 Great Castertone, 2 Rainham, 3 Neslé-Hodeng, 4 Macquenoise, 5 Hunneberg, 6 Samson, 7 Aachen, 8 Köln, 9 10 Krefeld-Gellep, 11 Brühl, 12 Bonn (Krausfeld), 13 Eichloch (Rommersheim), 14 Réhom, 15 Trier, 16 Trier, 17 Bingerbrük, 18 Wiesbaden-Igstadt, 19 Wiesbaden, 20 Worms (Maria Munster), 21 Heidelberg, 22 Palatin Hill (Rome), 23 Nocera Umbra, 24 CastelloTrosino, 25 Kranj, 26 Skadberg, 27 Høien, 28 Lærkenfeldt, 29 Sperrestrup, 30 Himlingøje, 31 Österhvarf, 32 Soukainen, 33 Nikutowen/Nikutowo (Evison 1975)* * I am indebted to Jacek Andrzeiowski and Maxim Levada for their help with this map.

345 Roman Import or Germanic Influence? On the Problem of Sarmatian-Germanic Contacts in the Crimea According to Glass Vessels Account 345 The excavation of glass-making workshop in Cologne discovered a beaker with cylindrical body, high truncated conical stem, soldered ovals, and single horizontal line below the lip, dated by Otto Doppelfeld to the third and fourth centuries (Doppelfeld 1966: fig. 132). A vessel of similar form, with high stem, of Roman production was discovered in 1842 in the municipality of Thiétreville on the northern France is kept in the museum of Ruan (La verrerie 1922: 198, fig. 256). Another remarkable find is also worth mentioning investigation in the hollow of Kara- Agach, Akmola uyezd (district), Kazakhstan excavated a barrow with rich nomadic woman s grave. There was glass vessel of pale green glass, with the body ornamented with soldered zigzag and straight strips of glass and big vertical ovals. The excavator, engineer Aleksandr Kozyrev, was not a specialist, so he dated the grave to the ninth century relating it with the Turks (Козырев 1905: 36). Having studied the ornaments from the barrow, Kira Skalon has attributed the assemblage to the late fifth century AD (Скалон 1962: 43). When developing his chronology of early mediaeval assemblages in the Central Asian steppes, Ambroz has dated the woman s grave in Kara-Agach to the sixth or seventh century (Амброз 1981: 17 18, рис ). Kropotkin was the first one who paid attention to the glass vessel dating it to the first half of the fifth century AD (Кропоткин 1970: 30, рис ). Irina Zasetskaya has correlated conical beakers and goblets with soldered threads from Kara-Agach barrow and disturbed Hunnic burial at modern Novaya Mayachka village with goblets discovered in South Korea and has dated these finds to the fifth century AD (Засецкая 1994: 103). In Boris Magomedov s opinion, beakers with soldered ovals in the Chernyakhov area imitate precious truncated conical beakers with cut oval medallions and dedicative inscriptions (Магомедов 2001: 66). One vessel of the kind originates from the cemetery of Shtilling, Denmark, possessing inscription Π Ι Ε Ζ Η Σ Α Ι Σ Κ Α Λ Ω Σ. Straume has dated it to phase С3 D1 (Straume 1987: 121). There are two more conical beakers with analogous inscription: one is a bit smaller, with low stem, from Vorning, Denmark (Straume 1987: 123), and another from Tu, Norway, with blue glass ornamentation (fig. 3. 7) (Straume 1987: ). Both vessels were discovered out of assemblages. Conical beaker of thick semi-transparent glass of yellowish tint was discovered in the cemetery of Mălăieşti, Moldova; on top body there is Greek inscription between horizontal incised lines: П I E Z H C A I C A Е I, with big ovals below, separated with convex cordon on projecting thick bottom (Федоров 1960: 286, рис. 11). Kropotkin has dated this find to the fourth century AD suggesting its eastern Mediterranean origin (Кропоткин 1970: 31, рис ). In their single chronological system for the cultures of Eastern Europe and Caucasus in the Gothic invasions period, Gey and Bazahn have attributed the Mălăieşti beaker to the fifth period (AD ) of the Chernyakhov culture (Гей, Бажан 1997: 45). Gavritukhin has mentioned one more conical beaker from the cemetery of Piwonice, Poland, dating it to the first half of the fifth century (Гавритухин 1999: 59). Ekholm has considered beakers with cut medallions and inscriptions from Scandinavia to be imitations of some vessels of Eastern Roman production still ill-known outside Denmark and Norway (Ekholm 1963: 34). In Berta Stjernquist s opinion, the place of manufacture of conical beakers with cut medallions and Greek inscriptions should be searched for in Greco-Roman centres of southeastern Europe as it is possibly indicated by the finds of similar vessels in Moldavia and Romania (Stjernquist 2004: ). There they concentrate between the Prut and Siret rivers and do not step out the Chernyakhov culture area and date back to the second half of the fourth and early fifth century by analogies (Pánczál, Dobos 2007: 74). Symonovich has written of almost total absence of analogous vessels in western Roman workshops Greco-Roman cities in the northern Black Sea and Mediterranean areas to suppose their local manufacture. In his opinion, the glass-making workshop discovered in Komarovo probably evidences for the resettlement of few fugitive glass-makers to the areas occupied by the Chernyakhov tribes in the early Great Migration periods to start glass-making there (Сымонович 1966: 109). The distribution area of glass conical beakers with soldered oval-like threads coincides with that of beakers with cut medallions to a large extent. If one agrees that cups with dedicatory Greek inscriptions were produced in Greco-Roman centres, the latter were most likely located in the imperial frontier on the Danube and Rhine and oriented on the customers living outside the empire. This idea is supported by almost complete absence of such vessels in the Roman imperial territory. Aleksander Bursche has pointed out that glass vessels from phase В2 С1b in Scandinavia, the Wielbark culture area, and Western Baltic culture, were im-

346 346 Stanislav SHABANOV ported from glass-making workshops in Cologne or Trier. Glass beakers of types Kowalk and Lugi appeared in the first half of the fourth century; according to Bursche, they evidenced the contacts of the populations of Northern and Central Europe with the Black Sea area (Bursche 1996: 35). The version of their production in the northern Black Sea area is still not able to find the confirmation (Shabanov 2011: 213). There also is no reasons for conclusion about the glass-making in the barbarian environment. Glass vessels ornamented with soldered threads in the form of large ovals initially were produced in workshops in the Apennine peninsula in the early Roman period. As Isings has put it, their finds concentrate in northern Italy and southern Switzerland where these vessels were probably produced in the first half of the second century AD (Isings 1957: 48). Tumblers with conical or cylindrical body and ring-foot are blown with applied oval decoration were distributed in the Roman imperial territory and outside it in the second and third centuries. They are discovered in Britain (Allen 1998: 36, fig ; Cool, Price 1995: 71, fig. 5. 5), Italy and Switzerland (Isings 1957: 47 78), Poland (Natuniewicz- Sekuła, Okulicz-Kozaryn 2008; 259, fig ; Eggers 1951: 178), Hungary (Barkoczi 1988: Taf. XII. 127), and Romania (Ionita 2000: fig ). The fourth century finds of such vessels are few (Isings 1957: 48). Conical vessels that could be used as drinking vessels and lighting lamps became popular in that period high and developed further later on. These vessels were ornamented in different ways and treated cold or hot. Although we can suppose that the ornamentation of glass vessels from the Roman period with applied ovals was developed further as well, it is still a guessing unless reliable evidences for the production of such vessels in Greco-Roman centres occurs. Most likely, the two glass beakers were brought to the Crimean foothill area simultaneously, because they are identical and rare. I cannot give an unequivocal answer to the question how so rare vessels found their way to the Crimean foothill area, but only suggest a few possible versions. First, the answer could be supplied by the goods discovered with the beaker in vault no. 485 in Neyzats cemetery. A person buried there probably got the vessel together with T-shaped brooch and glass cup with blue drops from Chersonesos or Bosporos, where it had been brought from Roman provinces on the Danube or in Gallia. Various glass vessels by Western Roman artisans occurred in Bosporos in the first century AD, with their number increasing in the third and fourth centuries (Кунина 1984: 160). Second, we should not omit the fact that geographically close finds of these vessels are in the Chernyakhov culture area. The Germanic tribes came to the Crimea about the mid-third century AD (Храпунов 2004: 146) in relation to the so-called Gothic invasions (Айбабин 1999: 26 30) accounted to by written sources (Хайрединова 1995: ). Archaeological data suggest that the Germans appeared in the Crimea as early as the first half of the third century AD (Власов 1999: 64, 66; Храпунов 2004: 159). There are hypotheses that the Germanic migration to Bosporos started in the second century AD (Шаров 2010: 282). Although from the mid-third tot he fourth century there were cremation cemeteries related to the Germans in the south-western Crimea and in its southern coast (Chyornaya River, Sovkhoz no. 10, Ay-Todor, and Chatyr-Dag), none of them possessed complex of features attributable to this or that archaeological culture. We should probably relate the appearance of two glass beakers to the migration of a group of Germans to the Crimean foothill area in the fourth century or to the contacts of Sarmatian-Alanic population with the Chernyakhov people, which are evidenced by researches of glass vessels (Shabanov 2011), hand-made vessels (Власов 2000), ornaments (Стоянова 2004), and weapons (Храпунов 2010). Discovered in cemeteries of Suvorovo and Neyzats glass vessels of Rau s types Frunzovka and Sigersted-Ganzkow do not find exact analogies outside the Crimea, though the most number of vessels with similar ornamentation occur in the entire European territories including the Crimea. Fragmented glass hemispherical bowls with two lines of fine cuts (fig ) were discovered in pit grave no. 25 in Suvorovo cemetery from the fourth century AD (Зайцев 1997: 110) and grave no. 22 (56) of Chyornaya River cemetery in the south-western Crimea (Шабанов 2013: рис ). Vault no. 1 in Pereval noye cemetery contained ring-footed tumbler with two lines of polished cuts, dated by Aleksandr Puzdrovskiy to the last quarter of the third century AD (Пуздровский 2007: 196, рис ). This vessel finds close analogy in Marseilles, southern France, from the third or fourth century AD (Foy 1998: 98, fig ). Both vessels have walls thinner than over artefacts of that type, and ring-foot. If Puzdrovskiy s chronology is correct, the Prevel noye find is perhaps the earli-

347 Roman Import or Germanic Influence? On the Problem of Sarmatian-Germanic Contacts in the Crimea According to Glass Vessels Account est among the vessels of the so-called Cologne type in the foothills of the Crimea. In Roman Britain, bowls with cut decoration of strokes and geometric figures are known in the second and third centuries AD (Allen 1998: 44 45, fig ), in Semyonvoka barrow cemetery in the Dniester area in the second and third centuries AD (Симоненко 2012: 279, рис ; Гросу 1990: рис ), on the Balkans in the cemetery of Iacobeni close to Iaşi, Romania (Ionita : Abb.1), in Tyras (Сон 1988: рис. 2), in the cemetery of Sirenevaya Bay in the Azov Sea area (Масленников 2000: 144, табл. III. 8), and in Tanais in the fourth century AD (Яценко 2007: , табл ). In Nina Kunina s opinion, cut glass vessels of the type in the northern Black Sea area are most likely produced in Colonia Agrippina (modern Cologne): local workshops produced similar vessels but in small quantity and for the local market only (Кунина 1984: ). Now we can only infer that ornamented bowls and phialae were made in glass-making workshop discovered by the excavations of the Late Scythian fortified settlement of Al ma-kermen in the south-western Crimea (fig. 11). 4 The excavator Tat yana Vysotskaya has related to the Roman legionaries who occupied the former barbarian settlement area in the second and third centuries AD. Among the fragments of few dozens glass vessels there were those with cut ovals, strokes, and geometric design (Висотська 1964: рис. 9). Yuliya Shchapova has made their morphological and typological analysis leading her to the conclusion that the Al ma-kermen workshop contained not only locally made vessels but also imports from other areas. Shchapova has pointed out that the absence of industrial wastes amidst the cold-treated vessels suggests their non-local origin (Щапова 1983: ). I can hypothesize that cut glass vessels appeared in the Crimea in the late third and fourth centuries together with other artefacts related with the people of Germanic circle: Eggers 230 beakers, amber and bronze ornaments, gray-clay vessels, and weapons. Especially interesting in the context of this paper are the finds of two glass rhyta, a rare type 347 in the Crimea, from the cemeteries of Sovkhoz no. 10 and Skalistoye III in the south-western Crimea. 5 Isings has called these glass vessels drinking horns singling them out as form 113. In her opinion, they are similar to conical beakers from the Late Roman period (her forms 106 and 109), from which they could borrow ornamentation techniques (Isings 1957: 141). Vera Evison has collected the finds of glass drinking horns from the Roman and early Mediaeval periods in Western and Central Europe to divide them into four types. The vessel from Sovkhoz no. 10 could be selected to type I (plain drinking horns, non-ornamented or with soldered colourless or colour thread), from the late fourth century, distributed in the Rhine area and Scandinavia (Evison 1975: 74 77). There is a vessel from Himlingøie, Seeland, Denmark with ornamentation similar to the Skalistoye III drinking horn. Eggers has dated that find to phase С2 (AD ) interpreting it as Roman import (Eggers 1951: 181). In Eastern Europe outside the Crimea, we know only one glass drinking horn (fig. 12.I 4), in Shapka cemetery in Abkhazia (Воронов 1975: 76 77). No common opinion exists for the chronology of the latter vessel. Yuriy Voronov has dated it to the sixth century as the earliest (Воронов 1973: 75), and Nina Sorokina, who has written a special paper discussing glass vessels from Tsibilium, suggests their earlier chronology, of the fourth or fifth century AD (Сорокина 1979: 63), relating their appearance in Tsibilium valley with the Romanization of local Apsilian population or their contacts with the residents of Roman forts in the eastern Black Sea area (Сорокина 1979: 63 64). Geographical distribution of glass drinking horns allows the suggestion that they were brought to the Crimea either from western Roman provinces (where they were probably produced) or from the area of free Germanic tribes. No doubt, drinking horns were not popular among the population of ancient Greek poleis in the northern Black Sea area throughout the entire Roman period, and the finds of such vessels in the Roman empire are not numerous as well. Evison s account makes it reasonable to conjecture the highest demand on these 4 The nearest glass-making workshop outside of the Crimea producing glass bowls with cut design is uncovered in the ancient town of Tanais (Алексеева, Арсеньева 1966: 188; Шелов 1972: 101), but it worked for a short time as long as the mid-third century, i. e. before the mentioned vessels occurred in the Crimean foothill area. 5 It is worth mentioning that so far we know the finds of three broken drinking horns from the sanctuary at Gurzufskoye Sedlo mountain pass on the southern coast of the Crimea. They were uncovered in a layer from the Hellenistic period, being made of polychrome glass, with possible chronology within the second or first century BC (Новиченкова-Лукичёва 2010: ).

348 348 Stanislav SHABANOV vessels among the Romanised barbarian elite on the Rhine, so all drinking horns but a few exceptions occurred in the Barbaricum in the late antique period and early Middle Ages. Since Crimean graves with drinking horns were not accompanied with other Germanic artefacts, one cannot be sure that they were brought there by Germans, and a century-long chronological gap between two groups of finds inclines me to infer that they were brought there from different places: the Chernyakhov culture area and the Roman provinces via Chersonesos and Bosporos. Bibliography Allen D. Roman Glass in Britain. Princes Risborough, Ancient Glass Research along the Silk Road. New Jersey, Andersen S. F. The Tylos Period Burials in Bahrain. 1. The Glass and Pottery Vessels. Aarhus, Barkoczi L. Pannonische Glassfunde in Ungarn // AAH IX. Bursche A. Contacts between the Late Roman Empire and North-Central Europe // The Antiquaries Journal Cool H. E. M., Price J. Roman vessel glass from excavations in Colchester, // Colchester Archeological Report Doppelfeld O. Römisches und Fränkisches Glas in Köln. Köln, Eggers H. J. Der römische Import in freien Germanien. Hamburg, Ekholm G. Orientalische Gläser in Skandinavien Während der Kaiser- und Fruhen Merowigerzeit // Acta Archaeologica. XXVII. København, Ekholm G. Scandinavian Glass Vessels of Oriental Origin from the First to the Sixth Century // JGS V. Evison V. Germanic Glass Drinking Horns // JGS XVII. Florescu R., Daicoviciu H., Roşu L. Dicţionar enciclopedic de artă veche a României. Bucureşti, Foy D. La Verre // Fouilles à Marseille Les mobiliers (I er VII er siècles ap. J.-C.) / Etudes massaliètes. 5. Paris, Gavritukhin I. Cut Glass Beakers within the Context of Studies in the Connections between the South of Eas tern Europe and Scandinavia in the Late Period of Roman Influence and the Great Migration Periods // Inter Ambo Maria: Contacts between Scandinavia and the Crimea in the Roman Period. Kristiansand; Simferopol, Harden D. Roman Glass from Karanis Found by the University Michigan Archaeological Expedition in Egypt, Michigan, Holland I. Sustaining Life. Vessel import to Norway in the first millennium AD. Stavanger, Ioniţă I. Eine Glasschale mit eingeritztem Wabenmuster und Facettenschliff von Iacobeni (Kr. Iaşi) // Dacia XXXVII XXXIX. Ioniţă I. Vase de sticlă Romane la Dacii liberi (secolele II III p. Chr) // Istros X. Isings C. Roman Glass from Dated Finds. Groningen; Djakarta, Khrapunov I. N. The Vault with Openwork Plaque from the Cemetery of Neyzats in the Crimea // The Turbulent Epoch. I. Lublin, Khrapunov I. The Northern Barbarians in the Crimea: a history of the investigation // Inter Ambo Maria: Contacts between Scandinavia and the Crimea in the Roman Period. Kristiansand; Simferopol, La verrerie en Gaule sous l Empire romain. Paris, Lazar I. The Roman Glass of Slovenia. Ljubljana, Lund Hansen U. Römischer Import im Norden // Nordiske Fortidsminder. Serie B Lund Hansen U. Contacts during the c. AD between South Scandinavia and the Black Sea Illustrated by Late Roman Glass and Jewellery // Inter Ambo Maria: contacts between Scandinavia and Crimea in the Roman period (Abstracts). Simferopol, Mączyńska M. The Early Medieval Necropolis of Almalyk-Dere near the Foot of Mangup in the South-West Crimea // Inter Ambo Maria: contacts between Scandinavia and Crimea in the Roman period (Abstracts). Simferopol, 2010.

349 Roman Import or Germanic Influence? On the Problem of Sarmatian-Germanic Contacts in the Crimea According to Glass Vessels Account 349 Natuniewicz-Sekuła M., Okulicz-Kozaryn J. Two richly furnished graves with Roman imports from the cemetery at Weklice, site 7, Elblag commune (Poland) // Germania Jahrgang Hallband. Pánczál S., Dobos A. Facet cut Glass Vessels of the Late 3 rd to 5 th century AD. Analysis of Finds from North Danubian Romania // Funerary offerings and votive depositions in Europe s 1 st Millennium AD. Cultural artefacts and local identities. Cluj-Napoca, Rau G. Körpergräber mit Glasbeigaben des 4. nachchristlichen Jachrunderts im Oder-Weichel-Raum // APA Shabanov S. Glass Beakers with Polished Ovals from the Late Roman Period in the Crimea // Inter Ambo Maria: Contacts between Scandinavia and the Crimea in the Roman Period. Kristiansand; Simferopol, Stjernquist B. Magnificent Glass Bowl from Uppåkra // Continuity for Centuries. A ceremonial building and its context at Uppåkra, southern Sweden / Uppåkrastudier. 10. Lund, Straume E. Facettslipte glass i nordiske graver fra 4. og 5. arhundrede e. Kr. Oslo, Straume E. Glaeser mit Facettenschliff aus skandinavischen Graeben des 4. Und 5. Jahrhunderts n. Chr. Oslo, Thomas S. Studien zu den germanischen Kämmen der römischen Kaiserzeit // Arbeits- und Forschungsberichte zur sachsischen Bodendenkmalpflege Айбабин А. И. Этническая история ранневизантийского Крыма. Симферополь, Айбабин А. И., Юрочкин В. Ю. Могильник «Баклинский овраг» (по материалам раскопок гг.) // Проблемы археологии древнего и средневекового Крыма. Симферополь, Алексеева Е. М., Арсеньева Т. М. Стеклоделие Танаиса // СА Амброз А. К. Фибулы юга европейской части СССР / САИ. Д М., Амброз А. К. Восточноевропейские и среднеазиатские степи V первой половины VIII вв. // Степи Евразии в эпоху Средневековья. М., 1981 Богданова Н. А., Гущина И. И., Лобода И. И. Могильник Скалистое-III в Юго-Западном Крыму (I III вв.) // СА Власов В. П. О появлении некоторых форм лепной керамики на позднескифских городищах Крыма в III в. н. э. // Проблемы скифо-сарматской археологии Северного Причерноморья. Запорожье, Висотська Т. М. Про виробництво скла в пiзньоантичному Криму // Археологiя XVI. Власов В. П. Трёхручные сосуды из могильника Дружное // МАИЭТ VII. Воронов Ю. Н. К истории экономических связей Апсилии в IV VII вв. Привозная стеклянная посуда из Цебельды // КСИА Воронов Ю. Н. Тайна Цебельдинской долины. М., Гавритухин И. O. Хронологические индикаторы финала Черняховской культуры // Сто лет черняховской культуре. К., Гавритухин И. О. Финал черняховской культуры // Восточная Европа в середине I тысячелетия н. э. / РСМ. 9. М., Гей О. А., Бажан И. А. Хронология эпохи «готских» походов (на территории Восточной Европы и Кавказа). М., Голофаст Л. А. Стекло ранневизантийского Херсонеса // МАИЭТ VIII. Гошкевич В. И. Древние городища по берегам низового Днепра // ИАК Гросу В. И. Хронология памятников сарматской культуры Днестровско-Прутского междуречья. Кишинёв, Гудкова А. В., Фокеев М. М. Земледельцы и кочевники в низовьях Дуная I IV вв. н. э. К., Зайцев Ю. П. Охранные исследования в Симферопольском, Белогорском и Бахчисарайском районах // Археологические исследования в Крыму Симферополь, Засецкая И. П. Культура кочевников южнорусских степей в гуннскую эпоху (конец IV V вв.). М., Козырев А. А. Раскопка кургана в урочище Кара-Агач Акмолинского уезда // ИАК Кропоткин В. В. Римские импортные изделия в восточной Европе (II в. до н. э. V в. н. э.) / САИ. Д М., Кунина Н. З. К вопросу о западном импорте стекла на Боспоре // ТГЭ XXIV. Магомедов Б. Черняховская культура. Проблема этноса. Lublin, 2001.

350 350 Stanislav SHABANOV Масленников А. А. Грунтовые некрополи сельских поселений Караларского побережья (Восточный Крым) первых веков н. э. // ДБ Никитина Г. Ф. Анализ археологических источников могильника Черняховской культуры у села Оселивка. М., Новиченкова-Лукичёва К. В. Три стеклянных полихромных ритона из эллинистического времени из святилища Южной Таврики // БИ XXIII. Петраускас О. В., Пастернак В. В. Скляні посудини могильника черняхівської культури Велика Бугаївка в Середньому Подніпров ї // Археологія Пуздровский А. Е. Крымская Скифия. III в. до н. э. III в. н. э. Погребальные памятники. Симферополь, Рикман Э. А. Вопросы датировки импортных вещей в памятниках племен черняховской культуры Днестровско- Прутского междуречья // СА Рикман Э. А. Этническая история населения Поднестровья и прилегающего Подунавья в первых веках нашей эры. М., Росохацкий А. А. Стеклянные кубки из памятников черняховской культуры степной зоны междуречья Днестра и Дуная // Новые исследования по археологии Северного Причерноморья. К., Румянцева О. С. Свидетельства стеклоделия в Черноморском регионе и на Днестре в свете новых данных // Черняхівська культура. Актуальні проблеми досліджень. До 40-річчя археологічної експедиції НПУ ім. М. П. Драгоманова (Тези доповiдей). К., Симоненко А. В. Стеклянные и фаянсовые сосуды из сарматских погребений Украины // Золото, конь и человек. Киев, Скалон К. М. Изображение дракона в искусстве IV V веков // СГЭ XXII. Смішко М. Ю. Поселення III IV ст. н. е. із слідами скляного виробництва біля с. Комарів, Чернівецької області // Матеріали і дослідження з археології Прикарпаття і Волині Сон Н. О. Скляні посудини з Тіри // Археологія Сорокина Н. П. Стекло из раскопок Пантикапея // МИА Сорокина Н. П. Позднеантичное стекло из Ольвии // Художественная культура и археология античного мира. М., Сорокина Н. П. Стеклянные сосуды IV V вв. и хронология цебельдинских могильников // КСИА Стоянова А. А. Бусы и подвески из могильника Нейзац (по материалам раскопок гг.) // БИ V. Сымонович Э. А. Стеклянная посуда середины I тысячелетия нашей эры с Нижнего Дона // КСИА Сымонович Э. А. Стеклянный кубок с надписью из-под Одессы // ВДИ Сымонович Э. А. Стеклянная посуда из поднепровско-причерноморских памятников черняховской культуры // СА Сымонович Э. А., Кравченко Н. М. Погребальные обряды племён черняховской культуры / САИ. Д М., Фёдоров Г. Б. Малаештский могильник // МИА Хайрединова Э. А. Боспор и морские походы варваров второй половины III в. н. э. // МАИЭТ IV. Храпунов И. Н. Могильник Дружное (III IV вв. нашей эры). Lublin, Храпунов И. Н. Новые данные о сармато-германских контактах в Крыму (по материалам раскопок могильника Нейзац) // БИ III. Храпунов И. Н. Подбойная могила позднеримского времени из некрополя Нейзац // ХСб XIII. Храпунов И. Н. Этническая история Крыма в раннем железном веке // БИ VI. Храпунов И. Н. Склеп IV в. н. э. из могильника Нейзац // ПИФК XXI. Храпунов И. Н. Оружие из могильника Нейзац // Terra Barbarica. Łódź; Warszawa, Храпунов И. Н. Некоторые итоги исследований могильника Нейзац // Исследования могильника Нейзац. Симферополь, 2011.

351 Roman Import or Germanic Influence? On the Problem of Sarmatian-Germanic Contacts in the Crimea According to Glass Vessels Account 351 Храпунов И. Н., Мульд С. А. Склепы с погребениями III в. н. э. из могильника Нейзац // БИ VII. Храпунов И. Н., Храпунов Н. И. Склеп с захоронениями IV в. н. э. из могильника Дружное в Крыму // ХСб Х. Шабанов С. Б. Стеклянные сосуды из могильника Нейзац (по материалам раскопок гг.) // МАИЭТ XVII. Шабанов С. Б. Стеклянные сосуды из раскопок Инкерманского и Чернореченского могильников (из коллекции Бахчисарайского заповедника) // Крым в сарматскую эпоху (II в. до н. э. IV в. н. э.). I. Симферополь; Бахчисарай, Шаров О. В. Данные письменных и археологических источников о появлении германцев на Боспоре // Stratum plus Шелов Д. Б. Танаис и Нижний Дон в первые века н. э. М., Щапова Ю. Л. Мастерская по производству стекла у с. Комарово III IV вв. // СА Щапова Ю. Л. Очерки истории древнего стеклоделия. М., Щукин М. Б. Готский путь (готы, Рим и черняховская культура). СПб., Юрочкин В. Ю., Труфанов А. А. Позднеантичный погребальный комплекс в низовьях реки Качи // ХСб XII. Яценко Е. Г. Коллекция стеклянных изделий из Танаиса IV V вв. н. э. // ДБ Станислав ШАБАНОВ Римский импорт или германское влияние? К вопросу о сармато-германских контактах в Крыму по материалам стеклянной посуды Резюме Ещё в XIX в. исследователи пытались разрешить проблемы, связанные с появлением и расселением на территории Крыма германских племён. Первоначально они опирались на данные письменных источников. Только в XX в. были сделаны значительные шаги в изучении германского присутствия в Крыму в позднеримское время, поскольку в распоряжение исследователей попали материалы археологических раскопок таких памятников как Ай-Тодор, Чёрная речка, Совхоз 10, Друж ное и Нейзац. Предметами специального исследования ста новились украшения, керамика и оружие, най денные в могильниках предгорного и горно го Крыма. Часть находок стеклянных сосудов исследователи традиционно связывают с по явлением представителей племен гер манско го круга, другие вполне могли быть элемен та ми римского импорта. К первым можно от нес ти кубки со шлифованными овалами (типы «Eg gers 223», «Eggers 230», «Straume VI» и «Stra ume VII, series A»), ко вторым стеклянные со су ды позднеримского времени с врезны ми ли ния ми и геометрическим орнаментом, най ден ные не только в Барбарикуме, но и на терри то рии Римской империи (тип «Eggers 195», а также «Sigersted», «Ganzkow» и «Frunzovka» по Г. Рау). Отдельные категории стеклянной по суды, например, ритоны («рога для питья»), ра нее вообще не учитывались в качестве ин дикаторов германского присутствия. В данной статье подробно рассматривают ся отдельные ка тегории стеклянных сосудов и то, каким образом они могли попасть в крымские предгорья.

352 352 Konstantin SKVORTSOV The Amber Coast Masters : some observations on rich burials in the Sambian-Natangian culture ca. AD 500 Before we enter the core of the problem, I would like to mention the reason that makes me to write this paper: the finds of new burial assemblages in the historical territories of Sambia and Natangia (modern Kaliningrad oblast, i. e. administrative area). Some of them could be the socalled princely graves from the Great Migration period (late fifth and early sixth centuries). This paper suggests a brief survey of these new finds presenting them to the reader and addresses some topical questions concerning the social structure of the Aestic 1 society ca. AD 500. It will not provide a deep insight into the analysis and interpretation of the new material because we plan to make a large and detailed publication introducing the results of these researches into the scholarly circulation in the future. More than 130-year-long archaeological investigations in the northern part of former Eastern Prussia found and studied only two burials from ca. AD 500 attributed to the so-called Fürstengrab, warlord graves: those no. 1 and 4 in Warnikam cemetery on the border of historical lands of Natangia and Warmia, discovered by Richard Klebs in summer of 1877 (Klebs 1878; Tischler, Kemke 1902: 41 42). This flat cemetery is located near the modern settlement of Pervomayskoye in the Bagrationovsk district of the Kaliningrad oblast (former Warnikam, Kreis [district] Heiligenbeil). Two-tier urnless cremation grave no. 1 contained the following grave goods (Tischler, Kemke 1902: 41 42, taf. XIII, XV. 9, 10; Hilberg 2009: ) (fig. 1. a.): silver crossbow brooch with cross-like ending of the foot-plate, of type Skvortsov-II (Скворцов 2010a: 41 42) (fig. 1. a. 3); silver neck-ring entwined of three wires 2 (fig. 1. a. 2); gold torqued bracelet, of variant Skvortsov 2.1 (Скворцов 2010a: 86 88) (fig. 1. a. 1); five small B-shaped silver panelled buckles; four small silver strap-ends; small silver corrugated ring (probably fibula spring stopper); single-edged iron scramasax with remains of wooden handle; fragments of stamped silver plates with ornamentation of sword s handle; three fragments of stamped silver plate in I Germanic animal style (Salin 1904: ; Haseloff 1981: ) (fig. 1. a. 5 7); small iron knife with remains of wooden handle; iron spear-head; two spurs. The low layer with horse s skeleton contained: iron horse bits with gilt bronze pieces, meeting with analogies among the goods of princely Gepidic grave Apahida-II (Quast 2007: 51, abb. 13) (fig. 1. a. 4); two gold cloisonné rosette-shaped strap-dividers with garnet insets (fig. 1. a. 9); 223 silver rivets; silver stamped plate with geometric and pearl ornamentation and rosettes (drinking horn fitting?) (fig. 1. a. 8); strapend; oval iron saddle-girth buckle; fragment of second bronze buckle. 1 In recent decades, the scholarship applies this conditional name to the population of the Sambian-Natangian culture from the Early Roman period to the Great Migration. 2 Vladimir Kulakov s publication of 1990 and his researches that followed confused the information about the goods in this burial, the neck-ring and bracelet in particular. Some of them state that three-wire neck-ring was of gold and smaller bracelet of silver, or even mention that the grave contained gold bracelet and neck-ring, though another bracelet was silver, etc. (Кулаков 1990: 64, 100, таб. VI. 9; Кулаков 1997a: 114; Кулаков 2007: 339). However, Otto Tischler in the first publication of this complex referred to 2 Armringe (XV9, XVI0), i. e. two bracelets, though the captions to the illustrative plate stated: Fig. 9, Silber, Fig. 10 Gold, tordiert, thus leaving no doubt concerning the number of artefacts and the metal they are made of (Tischler, Kemke 1902: 41, taf. XIII, XV. 9, 10).

353 Fig. 1. Some finds from the cemetery of Warnicam / Pervomayskoye. a some finds from grave no. 1: 1, 9 gold, almandines; 2 3, 5 8 silver; 4 gilt bronze; b some finds from grave no. 4: 1 2 silver; c stray find; gilt bronze

354 354 Konstantin SKVORTSOV Two-tier urnless cremation no. 4 contained the following grave goods (Tischler, Kemke 1902: 42, taf. XIII. 1) (fig. 1. b): silver stamped plate ornamented in I Germanic animal style (saddle fitting?) (fig. 1. b. 1, 2); 80 silver rivets; two buckles with rectangular frames; iron spear-head; heavily corroded horse-bits. The information of these graves is repeated by numerous archaeological volumes many times 3 (Klebs, 1878; Tischler, Kemke 1902: 41 42, taf. XIII, XV. 9, 10; Åberg 1919: 104, Abb. 149; 110, Abb. 154; Gaerte 1929: 290, Abb b; 240, 244, 245; Кулаков 1990: 64, 100, таб. VI. 8, 9; Nowakowski 1996: 41 42, Taf. 77, 79; Кулаков 1997; Kulakov 1997: ; Bitner-Wróblewska 2001: ; Nowakowski 2007: ; Кулаков 2007; ; Hilberg 2009: ). In early 2000s, some romantic colleagues even attempted to relate these assemblages with warlords Bruteno and Widewuto of Prussian legends (Кулаков 2004: 60). However, today these researches today are of only historical interest. It is worth mentioning that almost all the researchers of the social structure of the Sambian-Natangian (Dollkeim-Kovrovo) culture in the late stage of the Great Migration period have used graves no. 1 and 4 in Warnikam cemetery as the only source of information. Among much rare references are the horse-rider s grave in Warengen, Kreis Samland (Hilberg 2009: 331), a hoard of prestigious ornaments in Klein Ottern (Hilberg 2009: ), military elite burial in Ekritten, Kreis Samland / Vetrovo with parade bladed weapons with gold foil decoration (Hilberg 2009: Fundliste 22.5; Казанский 2010: 47), grave 9 in Tengen (Kreis Heiligenbeil), also with parade weapons in scabbards (Hilberg 2009: Abb. 9.11, Fundliste 22. 6), and some other finds (see: Hilberg 2009: Fundliste 22, Abb ). The information on the problem became even scanter after the disappearance of all the Warnicam finds in the last period of the Second World War. Present-day scholars can only use the pre-war publications with drawings and detailed descriptions of some finds from these graves, though others are only mentioned as a very few words. During recent 15 years, regular surveys and excavations by the Sambian-Natangian Expedition of the Institute of Archaeology of the Russian Academy of Sciences discovered a group of burial assemblages from the late Great Migration period, by many features comparable with the elite graves in the cemetery of Warnikam. They become the only material evidence of the ancestral nobility in the territory in question about AD 500. These new researches allow the analysis of the so-called chieftain graves in Sambia in other context, so let us turn to them. In 1997, protective investigations of flat cemetery of Kleinheide, Kreis Königsberg (fig. 2.a.), in the Gur yevsk district, Kaliningrad oblast by the Natangian Team of the Baltic Archaeological Expedition of the Institute of Archaeology of the Russian Academy of Sciences in cooperation with the Kaliningrad State University discovered grave assemblage no. 21, by all the features of its funeral rite and grave goods belonging to a privileged person (Скворцов 1998; 1999: 50; Кулаков 2003: ). Assemblage no. 21 is a collective cremation burial 4 in wooden box instead of urn (fig. 2. b, d), covered by thick multi-tiered stonework (fig. 2. c). The set of goods suggests the early sixth century as the date for the assemblage. The wooden box contained the following grave goods (fig. 2): five bi-conical hand-formed vessels of the same type, decorated with incised zigzag and four cordons on the rib (fig. 2.e.11, 12, 14 16); gold torqued bracelet (fig. 2. e. 1), which is almost the same as the find in the assemblage no. 1 in Warnicam cemetery; three knives with remains of wooden handles (fig. 2. e. 8, 9, 13); five bronze fibulae with cross-like ending of the foot-plate, of type II (Скворцов 2010a: 41 42). (fig. 2. e. 2 6); two iron battle knives with wooden handles, etc. (fig. 2. e. 7, 10). Although this grave assemblage has construction and funeral rite unusual for local antiquities, at the same time it finds multitude of analogies among the burials in southern Scandinavia where such funeral rite originated in the Bronze Age and existed to the Viking age (Stenberger 1977: Abb. 150, 3 Very often these burials are published together with a picture of bronze shield-like gilded ornamented appliqué (fig. 1. с), which some researchers without any argument attribute to grave no. 1 (Кулаков 2007: 355, рис ). 4 According to the preliminary results of Mariya Dobrovol skaya s anthropological research of the material from this grave, it contained the remains of at least three persons and bones of a dog, also cremated on funeral pyre. It should be noted that the tradition of putting dogs into burials is recorded in nobility graves in Central Europe and Scandinavia in the late Great Migration period and the Merovingian age (see for example: Öhman 1982).

355 The Amber Coast Masters : Some Observations on Rich Burials in the Sambian-Natangian Culture ca. AD Здесь широкоформатный рис размерами 480 х 297 мм Цветность листа 4+4 (В отдельном файле)

356 356 Konstantin SKVORTSOV Здесь рис размерами 480 х 297 мм Цветность листа (В отдельном файле)

357 The Amber Coast Masters : Some Observations on Rich Burials in the Sambian-Natangian Culture ca. AD ). Quite probably, the innovations in the funeral rite appeared in result of an outside impulse from Scandinavia. 5 The Sambian Expedition of the Institute of Archaeology of the Russian Academy of Sciences 2008 protective investigations of flat cemetery of Mitino / Stantau, Kreis Samland (fig. 3. a), discovered a partly disturbed grave with the goods dating from the late fifth and early sixth century, which in my point of view had likely been among the nobility graves (Скворцов 2010a: ; Скворцов 2010b: 95 96, ; Skvorzov, Pesch 2011: ). Assemblage no. 335 initially was a burial of rider and his horse, with the man located on the east, but it was unfortunately destroyed, so only fragmented hand-made ceramic vessel and a buckle fragment survived. Twin horses grave that accompanied the main burial remained undisturbed. There were elements of horse tack in anatomical order: thirty-two fragmented silver appliqués; three silver and one bronze belt-ends; two (?) fragmented silver strapdividers with stamped ornamentation including anthropomorphic pictures; five small bronze oval buckles; two cheek-pieces. In the north area of the pit, there were artefacts related to the horse tack, located in anatomical order on the horse s rump, and namely silver fittings of saddle pommel and cantle with stamped decoration (fig. 3. b). The finds of similar saddle parts are unique for Sambia, Natangia, and Eastern Prussia in general. The ornament of zoomorphic and geometric elements on the plates meets with partial analogies in decorative motifs of the artefacts in the same technique from Lithuania and Scandinavia (Schetelig 1912: 111, 113, fig. 253; Almgren, Nerman 1923: taf. 31, fig. 456; Åberg 1931: 54 55, Abb. 145; Vaškevičiūtė 1978: pav. 4; Andrzejowski 1991: 49 50; Kazakevičius 1993: ; Šimėnas 2006: 55, pav ). However, we should mention that Lithuanian colleagues tends to assume that zoomorphic and geometric motifs, particularly those on drinking horn fittings from Lithuania, are generally rare in the Baltic area, so the origins of 357 this decoration should be looked for in Scandinavia (Kazakevičius 1993: 136). The saddle fittings from the cemetery of Mitino are unique because of the image of warriors, small but well executed, with high particularization. So far, similar anthropomorphic images do not meet with analogies among the Baltic materials. General appearance of the figurines of warriors on saddle plates from Mitino bears much similarity with the iconography of Northern Germanic bracteates from the Great Migration period. These pendants were made mainly in southern Scandinavia and neighbouring areas of Northern Europe (Pesch 2007: ; Skvorzov, Pesch 2011: ) In the end of the field archaeological season of 2012, a team of the Sambian Expedition headed by the author of given research that was doing surveys in several districts of the Kaliningard oblast discovered two burials with rich grave goods, comparable with graves in Warnikam cemetery. Both assemblages could be interpreted as chieftains graves, also from the late fifth or early sixth century according the set of grave goods. The first find is made in the cemetery located in the Sambian peninsula, near modern settlement of Logvino (Klein Medenau, Kreis Samland). There was a heavily disturbed horse s burial. 6 The part of the grave that survived contained the elements of horse s headgear (fig. 4), particularly fragments of iron bits with reigns shaped like dragon s heads, with niello and hot gilding. Although similar dragon-head-shaped design of horse bits fitting occurred in grave no. 1 of Warnikam cemetery, as far as we can judge from the illustrations that survived, their style was less artistic. There also were silver brow-band pendant in I Germanic animal style, with niello and hot gilding (fig. 4. 1); three strapdividers in I Germanic animal style, with niello and hot gilding, and almandine insets (fig. 4. 2); bronze panelled buckle decorated with silver plate. The style and composition of this set of artefacts obviously suggest their dating to the late fifth and early sixth century; the single style of all the elements makes us to consider that they were pro- 5 In respect of history, these phenomena come to agreement with the Prussian chronicles legendary account of the arrival of Widewuto and Bruteno, legendary forefathers of the Prussians, who sailed from Denmark, to the Vistula bay in the early sixth century (Grunau 1876). Although this find does not prove the reality of these persons, it is a brilliant evidence of the contacts with Scandinavia about AD Most likely, this area was damaged in February or March 1945, when it was a battleground, so its territory houses numerous fighting pits, caponiers, and bunkers.

358 358 Konstantin SKVORTSOV Fig. 4. Some of the finds from the flat cemetery of Logvino. 1 2 gilt silver, niello, almandine duced as a single set. All the artefacts 7 in this complex were made in I Germanic animal style suggesting their immediate connection with their origin in Scandinavian territory. Such finds in the land of the Balts could only be imports; no doubts, this assemblage belonged to ancestral elite (Salin 1904: ; Haseloff 1981: ; Hilberg 2009: 331). In the same cemetery, we earlier discovered a ruined woman s grave from the Late Roman period with gold and silver artefacts, also attributed to a person of privileges in this society. It can be inferred from this fact the given site from the Roman period on the side of the Nel ma river, playing an important role in communication of the residents of the Sambian peninsula and the Vistula estuary since the Bronze Age, was created by the Sambian-Natangian people who had been taking active part in various inter-ethnic contacts. The luxury artefacts in graves from that period could be related to the processes recorded by other colleagues many times: the migration of the Sambian-Natangian population to the west, into the areas formerly occupied by the Wielbark people, resulting in the formation of a new contact zone closer to the Vistula in the final of the Great Migration period (Скворцов 2013). The last find of 2012 was made in the cemetery of Shosseynoye, 1.5 km south of the coast of Kaliningrad bay, close to the former manor Maulen, Kreis Königsberg, which does not exists now (Скворцов 2012: 9 12). There was a partly disturbed cremation burial of horseman, put into the grave within a wooden box. This assemblage was accompanied of at least five horses burials, 8 and also remains of the second person, probably a servant, within the horseman s grave (he had been the companion accompanying the master in the other world). This grave contained goods 7 Apart of the buckle, because a great deal of buckles of the type occurred in all the Sambian-Natangian culture sites from the late Great Migration period. 8 A great number of horses graves accompanying riders burials also indicated prestige in the Merovingian time (see for example: Ghenne-Dubois 1991: 55; on the horse s burials in royal graves see: Kazanski, Périn 2005: ; Périn, Kazanski 2007: 32 36).

359 The Amber Coast Masters : Some Observations on Rich Burials in the Sambian-Natangian Culture ca. AD Fig. 5. Some of the finds from the flat cemetery of Shosseynoye. 1 fragment of silver saddle fitting; 2, 4 gilt bronze; 3 gilt silver brooch; 5 gilt bronze pendant; 6 7 iron spear-heads of unique richness and variety (fig. 5), particularly silver gilt radiate-headed brooch of type Purda (Kowalski 1991; Kowalski 2000) (fig. 5. 3); two bronze panelled buckles; two bronze pincers; two iron spear-heads, absolutely the same as the spears discovered in assemblages of Warnikam cemetery. Spear-heads of the type are well-known in the fifth and sixth centuries in the Baltic Sea area, both in the Balts land and in bog offerings in Denmark (Hilberg 2009: 328) (fig. 5. 6, 7). There also were two amber beads, fragments of a few ceramic vessels, and a fragment of iron knife. The elements of horse tack included: numerous iron ringed horse-bits and silver rivets, and various bronze harness and girth buckles. One of the horses in this burial was especially richly equipped, includ-

360 360 Konstantin SKVORTSOV Fig Logvino; 2 Mitino; 3 Kleinheide; 4 Shosseynoye; 5 Warnikam ing: iron bits with gilt bronze and silver details of Judith Oexle s type I (Oexle 1992: 35 39); numerous silver rivets; two gilt bronze front pendant in I Germanic animal style (fig. 5. 5); four gilt bronze strap-dividers in I Germanic animal style (fig. 5. 2, 4); various saddle fittings of figural stamped silver plates in I Germanic animal style (fig. 5. 1). It is interesting to note that some fittings are made by the same artisan and the same tool (stamp), as the stamped plates from burials no. 1 and 4 in the cemetery of Warnikam, which is not a surprise since Shosseynoye cemetery is located only 21 km east of Warnikam, within the same section of land road from the Sambian peninsula to the Vistula estuary. Grave goods allow us to date this assemblage the same as the listed above sites, about AD , and the goods from these graves contain a great number of analogies to Scandinavian materials. All the complexes discovered in the last 15 years and graves no. 1 and 4 in Warnikam cemetery have several features in common, so we can attribute them to the heads of large families or clans that probably controlled most important routes allowing the local population to trade with neighbouring areas on the Baltic Sea, as well as with Scandinavia and Central Europe. Among these features are: All the mentioned assemblages are not isolated, as it is typical of noble tribal chiefs graves ( Fürstengrab ) in the Barbaricum, but placed in family cemeteries, surrounded by the burials of very different social strata including poor ones. The number of burials of interest is certainly small. All the cemeteries with such grave assemblages are located either near the places where amber was mined or on the main section of the trade route from Sambia to the Vistula estuary, thus continuing the tradition from the age of the Roman influence (fig. 6). The composition of the grave goods in these assemblages is more varied and richer than in other graves within the same cemetery. They always contain artefacts of precious metals and, together

361 The Amber Coast Masters : Some Observations on Rich Burials in the Sambian-Natangian Culture ca. AD 500 with locally made ware, imports from Germanic environment (applying techniques of hot gilding, niello, stamping, and cloisonné, with almandine insets, decoration in Germanic animal style, etc.). This case is not restricted to ornaments, but also includes horse harness details. Most part of imports are of Scandinavian origin. All the graves of the kind contained cremations, typical of the Sambian-Natangian culture in the late Great Migration period. Although their funeral rite is generally typical of the local culture, it still includes several elements of the Germanic circle. This way, in two instances human remains were put into grave within a wooden box, as it is especially understandable with grave no. 21 in Kleinheide. Men s burials were as a rule accompanied with graves of one and in some cases few horses (up to five animals or more), and often with remains of servants or slaves ( companions ). Therefore, we the new finds allow to enlarge our knowledge of the social features of the Sambian- Natangian culture. We can be more or less convenient that the group of the above-described burials belonged to local nobility, i. e. the heads of clans, representing ancestral aristocracy, rather than princes or kings of large tribal unions. It is still possible that there were some tribal chiefs among the Sambian-Natangian society from the late Great Migration period, and their graves would be discovered in the future in the inland of the Sambian peninsula (Skvorzov 2013). As an indirect evidence for that hypotheses could be interpreted Cassiodorus account of the Aestii s embassy with amber gifts to the Ostrogothic king Theodoric the Great about (Хенниг 1961: 60 61). This diplomatic mission was probably an evidence of some uniting element of power, like a tribal chief or tribal union, who could initiate such a mission. In this epistle, we should also pay attention to Cassiodorus reference to Tacitus Germania (Kolendo 1990: 95 97; Nowakowski 1995: 83). Archaeological finds suggest that amber remained a popular commodity in the Great Migration period. It coincides with the rich graves concentration at places of importance for amber transportation to the Vistula estuary (fig. 6). Large clans occupying land roads could also control coastwise routes along the coast 361 of Vistula bay and the seashore of Sambia (Кулаков 2007: 340; Казанский 2010: 46 48; Казанский, Мастыкова 2013). The importance of this trade area directed to the Vistula estuary is underlined by numerous hoards including the treasure of Młoteczno (former Hammersdorf, Kreis Heiligenbeil) (Кулаков 2003: ; Cieśliński 2010: ). We should also mention that not far from that place, in the Vistula estuary, the sixth century historian Jordan located the Vidivarii, a people gathered out of various tribes. The researchers have interpreted the Vidivarii as a mixed poly-ethnic unit of Gepids and Aestii (Иордан 1997: 64 65, 67, 80, 84). Perhaps the Vidivarii started the processes to result in the shaping of the new community of the Prussians. Some researchers have related the Vidiarii with the so-called Elbląg group from the same period (Godłowski 1981: ). All that happened against the background of crucial ethnocultural processes in the south-eastern Baltic Sea area and in neighbouring territories in the late fifth and early sixth century was related to the disappearance of the Wielbark culture in the early fifth century, that had been a buffer in the Sambian population s contacts with Central Europe and the Roman Empire. Simultaneously, the Wielbark community played the mediating role in the amber trade throughout its lifetime since ca. AD 1 to the early Migration period (Skvorzov 2013). These processes resulted in the appearance of the Elbląg and Olsztyn cultures groups. The late fifth and early sixth century archaeological materials indicate that the main source supplying the nobility of Sambia with artefacts of prestigious culture was located in Scandinavia. 10 The Sambian-Natangian population traditionally contacted with the population of the Scandinavian peninsula and the Baltic Sea islands by sea, and these connection had not been interrupted in previous historical periods (Lund Hansen 1988: , Abb. 2; Skvorzov 2013). This direction coincides with the source of imported ware in the Elbląg group, which key contacts were also with Scandinavia (Kontny, Okulicz-Kozaryn, Pietrzak 2011: ). On the contrary, the Masurian Lakeland materials (Olsztyn group) indicate permanent links to Southern Europe and minimal 9 However, there is an interpretation of that epistle as just a sample of correspondence invented by Cassiodorus. Anyway, no one calls this embassy, dated to by Jerzy Kolendo, impossible (Kolendo 1990: 95 97; Nowakowski 1995: 83). 10 On the Scandinavian-Baltic contacts in the Great Migration period see: Bitner-Wróblewska 2001.

362 362 Konstantin SKVORTSOV contacts with Scandinavia (Okulicz 1973: ; Hilberg 2009: 332). Prestigious finds in possession from nobility graves of the Sambian-Natangian culture, partly related to Scandinavia by origin, allow the one to trace Scandinavian warrior and chieftain elites influence on local elites. This influence was spread by maritime trade routes connecting the Scandinavians with the populations of the Vistula estuary and Sambian peninsula and their possible matrimonial relationship as well (Несман 1989: 21; Bitner-Wróblewska 2001: ; Казанский, Мастыкова 2013). Marriages with remote peoples formed a tool of politics, as it was brilliantly reflected in early mediaeval Germanic epos. Local tribes of Sambia and Natangia, located on the edge of the civilization, and particularly their nobility seem to participate in that heroic age and in historical events changing the entire ethnic and political map of Europe. Bibliography Åberg N. Ostpreussen in der Völkerwanderungszeit. Leipzig; Uppsala, Åberg N. Nordische Ornamentik in vorgeschichtlicher Zeit // Mannus Bibliothek. 47. Leipzig, Almgren O., Nerman B. Die ältere Eisenzeit Gotlands. Stockholm, Andrzejowski J. Okucia rogów do picia z młodszego okresu przedrzymskiego okresu wpływów rzymskich w Europe Środokowej i Północnej (Próba klasyfikacji analizy chronologiczno-terytorialnej) // Materialy Starożytne Wczesnośredniowieczne VI. Bitner-Wróblewska A. From Samland to Rogaland: East-West Connections in the Baltic Basin during the Early Migration Period. Warszawa, Cieśliński A. Kulturelle Veränderungen und Besiedlungsabläuffe in Gebiet der Wielbark-Kultur an Łyna, Pasłęka und Oberer Drwęca // Berliner Beiträge zur Vor- und Frühgeschichte. 17. Berlin, Gaerte W. Urgeschichte Ostpreußens. Königsberg, Ghenne-Dubois M.-J. Lez inhumations de chevaux et les trouvbialles de Saint-Brice // Les fouilles du quartier Saint- Brice à Turnai. 2. L environnement funéraire de la sépulture de Childéric. Louvnain-la-Neuve, Grunau S. Preussische Chronik. I/II. Leipzig, Godłowski K. Okres wędrówek ludów na Pomorzu // Pomorania Anitiqua X. Haseloff G. Die germanische Tierornamentik der Völkerwanderungszeit. I. Berlin; New York, Hilberg V. Masurische Bügelfibeln. Studien zu den Fernbeiziehungen der völkerwanderungszeitlichen Brandgräberfelder von Daumen und Kellaren // Daumen und Kellaren-Tumiany i Kielary. 2 / Schriften des Archaologischen Landesmuseums. 9. Neumünster, Kazakevičius V. Plinkagailio Kapinynas / Lietuvos Archeologija. 10. Vilnius, Kazanski M., Périn P. La tombe de Childéric: un tumulus oriental? // Mélanges Jean-Pierre Sodini / TM. 15. Paris, Kontny B, Okulicz-Kozaryn J, Pietrzak M. Nowinka, Site I. The cemetery from the late Migration Period in the northern Poland. Gdańsk;Warszawa, Klebs R. Über einen Goldfund in Natangen // SPÖG. XIX. Königsberg, Klebs R. Über das Grabfeld von Warnikam bei Ludwigsort // SPÖG. XXII. Königsberg, Kolendo J. Napływ bursztynu z północy na tereny Imperium Rzymskiego w I VI w. n. e. // Prace Muzeum Ziemi Kowalski J. Z badań nad chronologią okresu wędrówek ludów na ziemiach zachodniobałtyjskich (faza E) // Archeologia Bałtyjska. Olsztyn, Kowalski J. Chronologia grupy elbląskiej i olsztyńskiej kręgu zachodniobałtyjskiego (V VII w.) // Barbaricum. 6. Warszawa, Kulakov V. Gräber pruzzischer Stammesführer aus Warnikam // Eurasia Antiqua Lund Hansen U. Handelszentren der römischen Kaiserzeit und Völkerwanderungszeit in Dänemark // Trade and Exchange in Prehistory. Lund, 1988.

363 The Amber Coast Masters : Some Observations on Rich Burials in the Sambian-Natangian Culture ca. AD Nowakowski W. Od Galindai do Galinditae. Z badań nad pradziejami bałtyjskiego ludu z Pojezierza Mazurskiego // Barbaricum. 4. Warszawa, Nowakowski W. Das Samland in der römischen Kaiserzeit und seine Verbindungen mit dem römischen Reich und der barbarischen Welt // Veröffentlichungen des Vorgeschichtlichen Seminars Marburg. Sonderband. 10. Marburg, Nowakowski W. East Prussia as a bridge between Eastern and Western Europe: finds oft he 5th to 8th centuries // The Merovingian Period-Europe without Borders: Archaeology and history of the 5th to 8th centuries. München, Oexle J. Studien zu merowingerzeitlichem Pferdegeschirr am Beispiel der Trensen. Mainz, Öhman I. The Merovingian dogs from the boat-graves a Vendel // Vendel Period Studies. Stokholm, Okulicz J. Pradzieje ziem pruskich od pózdnego paleolitu do VII w. n. e. Wrozlaw, Périn P., Kazanski M. La tombe de Childéric, le Danube et la Méditerranée // Villes et Campagnes en Neustrie / Europe médiévale. 8. Montagnac, Pesch A. Die Goldbrakteaten der Völkerwanderungszeit Thema und Variation // Ergänzungsbände zum RGA. 36. Berlin; New York, Quast D. Zwischen Steppe, Barbaricum und Byzanz. Bemerkungen zu prunkvollem Reitzubehör des 5. Jahrhunderts n. Chr. // APA Salin B. Die Altgermanische Tierornamentik. Stockholm, Schetelig H. Vestlandske graver fra jernalderen // Bergens museums skrifter. Ny Række II/1. Šimėnas V. Etnokultūrinai Procesai Vakarų Lietuvoje pirmojo mūsų eros tūkstantmečio viduryje. Vilnius, Skvorzov K. N. The formation of patrimonial elite of Sambian-Natangain culture in Roman Period in the context of amber trade // Archaelogia Baltica /2 (in print). Skvorzov K. N. Pesch A. Krieger, Dicke Vögel und gehörnte Pferde? Ein Sattelbeschlag aus Mitino (obl. Kaliningrad) // Archäologisches Korrespondenzblatt. 41 / 3. Mainz, Stenberger M. Vorgeschichte Schwedens Aus dem Schwediscen ubersetzt von Hedda und Torsten Capelle. Berlin, Tischler O., Kemke H. Ostpreussische Altertümer aus der Zeit der grossen Gräberfelder nach Cristi Geburt. Königsberg, Vaškevičiūtė I. Gyvuliniai motyvai VI VII a. Žiemgalių papuošalų ornamentuose // Jaunųjų istorikų darbai. 2. Vilnius, Иордан. О происхождении и деяниях гетов (Gеtiса). 2e изд. (пер. Е. Ч. Скржинская). СПб., Казанский М. М. Скандинавская меховая торговля и «Восточный путь» в эпоху переселения народов // Stratum plus Казанский М. М., Мастыкова А. В. О морских контактах эстиев в эпоху Великого переселения народов // Археология Балтийского региона. СПб., 2013 (in print). Кулаков В. И. Древности пруссов VI XIII вв. / САИ. Г 1 9. М., Кулаков В. И. Варникам. Древности прусских вождей // Гiстарычна-Археалагiчны зборник Кулаков В. И. История Пруссии до 1283 года. М., Кулаков В. И. Боги Видевута // Letonica. 10. Rīga, Кулаков В. И. Самбия и Натангия // Восточная Европа в середине I тысячелетия н. э. / РСМ. 9. М Несман У. Этнос и связи на Балтике в V X веках нашей эры // Взаимодействие древних культур в бассейне Балтийского моря. Л., Скворцов К. Н. Отчёт о раскопках, проведённых Натангийским отрядом Балтийской археологической экспедиции Института археологии Российской академии наук в 1997 году. М., 1998 (Archive of the Institute of Archaeology of the Russian Academy of Sciences, collection R 1, no ). Скворцов К. Н. Работы Натангийского отряда Балтийской экспедиции // Археологические открытия 1997 года. М., Скворцов К. Н. Могильник Митино V XIX вв. (Калининградская область) // Материалы охранных археологических исследований. 1/15. М., 2010а. Скворцов К. Н. Могильник Митино V XIX вв. (Калининградская область) // Материалы охранных археологических исследований. 2/15. М., 2010b.

364 364 Konstantin SKVORTSOV Скворцов К. Н. Новые находки памятников римского времени на побережье Вислинского залива // Stratum plus Скворцов К. Н. Западные Балты и их соседи на Вислинском заливе в римское время // Археология Балтийского региона. СПб., 2013 (in print). Хенниг Р. Неведомые земли. 2. М., Константин СКВОРЦОВ «Владыки янтарного берега»: некоторые наблюдения по поводу богатых погребений самбийско-натангийской культуры рубежа V VI вв. Резюме Цель данной статьи дать краткий обзор новых находок погребальных комплексов на территориях исторических Самбии и Натангии (современная Калининградская область), которые по ряду признаков можно отнести к так называемым «вождеским» погребениям эпохи Великого переселения народов (конца V начала VI вв.). К данным находкам относятся: по гребение 21 могильника Кляйнхайде, об наруженное в ходе исследований 1997 г.; по гребение 335 могильника Митино, обна руженное в ходе исследований 2008 г.; разру шенное конское захоронение, об на руженное на могильнике Логвино в сезоне 2012 г.; по гребение, обнаруженное на могиль нике Шос сейное в сезоне 2012 г. В статье представлены краткие описания по гребального обряда и погребального инвен таря данных комплексов, а также поднят ряд актуальных вопросов, касающихся со циаль ной структуры общества эстиев рубежа V VI вв. Результаты новых исследований значи тельно расширяют и обогащают наши знания о знатных погребениях эпохи Великого пе реселения народов конца V начала VI вв. в ареале самбийско-натангийской культуры. Это связано с тем фактом, что до последнего вре мени археологическая наука имела в своем распоряжении исключительно один источ ник информации по данной теме ма тери алы исследований 1877 г. комплексов 1 и 4 могильника Варникам, которые утра че ны в ходе Второй мировой войны, причём сохранился лишь ряд довоенных, не всегда развёрнутых публикаций. Дальнейший ана лиз и полноценное введение в научный оборот результатов последних исследований, которым посвящена данная статья, позволит нам значительно продвинуться в изучении сам бийско-натангийской культуры рубежа V VI вв.

365 365 Frans-Arne STYLEGAR Regionality in Iron Age Norway. A study of archaeological cultures For Norwegian Iron Age archaeology, cultures in the classical sense do not seem to exist. Yet, looking at the first centuries after the birth of Christ, the Roman and Migration periods (AD 1 550), and focusing on burial rites alone, it is a fact that over large tracts of southeast Norway, cremation graves without any markings above ground are a significant type, while north of the Trondheim area there are hardly any cremations at all, and almost all graves are covered by barrows. And in between these two extreme opposites opposites both in a ritual sense and geographically are found other combinations with a local or regional distribution. Indeed, this is perhaps only what to expect in a country as varied topographically as Norway (fig. 1). Several mountain ranges and valleys divide the country into a number of nature-given regions: The Langfjellene divides the southern part of Norway into an eastern (Østlandet) and a western region (Vestlandet), while the former is subdivided into several mountain valleys as well as the lowland Oslofjord area and the clearly demarcated districts around the lakes Mjøsa, Randsfjorden and Tyrifjorden, and the latter into a number of fjord districts. Further north, the Trøndelag area is defined by the Dovre mountains to the south and the Helgeland mountains to the north. North of the Helgeland mountains is North Norway, the historical region of Hálogaland. Traditionally speaking, three different routes linked these major regions. Most importantly, a sea route, the so-called Leden, linked the different coastal districts. Over land, natural corridors linked Østlandet with Vestlandet through the valleys and across the Filefjellmountains, and Østlandet with Trøndelagthrough the valleys and across the Dovre mountains (Slomann 1956: 63). On the surface of it, this seems like an obvious place to look for archaeological cultures in the plural. But over the last century or so, and especially since the Second world war, the concept of culture in this sense has disappeared from the archaeological scholarly debate. What replaced it was initially a concern for economics, later on for politics, and more recently for belief systems and everything ideological. The latter would seem like fertile ground for the reintroduction of culture, but this has not been the case. There is only one real exception: archaeologists have been able to wrest Saami culture and Saami burial practices from the otherwise undifferentiated mass of cultural traditions in North Norway (Schanche 2000). Otherwise, culture is shunned, and so is often regionality, as well. This, I believe, is partly a result of the gradual demise of traditional, cultural historical archaology in Norway since the 1970s, but there are also less immediate reasons. In this paper I would like to concentrate on mortuary rites, or, to put it more precisely, the material remains of such rites. Considering the above situation, it is hardly coincidental that of the three synthesizing and superregional studies dealing with regional differences in the Roman and Migration periods, the most recent one was published in 1912 (Rygh 1869; Undset 1880; Schetelig 1912). Nationalism and the concept of regionality The immediate background lays in the early post-war years, when many scholars reacted against what they perceived to be the excesses of the Kossina school of Siedlungsarchäologie in the 1920s and 30s, and in particular its often too simple linkage between archaeological cultures and living cultures or ethnicities, as expressed in Kossina s famous dictum, sharply defined archaeological cultural areas correspond unquestionably with the areas of particular people or tribes (see: Klejn 1974; Klejn 1999). Instead, archaeologists turned to ecology, economics and social anthropology for models and theory, often with interesting results. Existing regional studies often have a political-economic starting point, for instance Bjørn Myhre s well-known Chieftains graves

366 366 Frans-Arne STYLEGAR Fig. 1. Norway has a very varied topography, characterized by mountain ranges, valleys, and deep fjords. Here a fjord landscape in western Norway (photo: Erik A. Drabløs / Wikimedia Commons) and chieftains territories from 1987, where he identifies regional overlordships or redistributive systems in Migration period SW Norway (fig. 2). He does mention Jordanes and some possible tribes mentioned in Scandza, but does not really go into the problem of attributing his regional systems to specific tribes or peoples (Myhre 1987). But in hindsight, it is hard to find many examples at all in the period between the two wars of a Norwegian archaeologist working within the paradigm of Siedlungsarchäologie in a strict sense. One could think of Th. Petersen in his study of the Trøndelag area in the Roman period, where he argued that immigration from present-day Denmark was behind changes in material culture and burial customs (Petersen 1930), Gutorm Gjessing s suggestion that the Migration period expansion in North Norway was carried by new immigrants from Rogaland in the southwestern part of the country (Gjessing 1929; Gjessing 1930), or Helge Gjessing s argument that Vest-Agder experienced an influx of new population-elements from the west in the fifth and sixth centuries (Gjessing 1925: 69). Otherwise, Haakon Shetelig (1920) probably comes closest, but his way of doing archaeology was still very different from Kossina s, even if both had a keen eye for regional patterns. For one, Shetelig never drew any political conclusions based on the distribution of archaeological cultures. It is no coincidence that these few examples deal with migration. Interestingly, there was hardly any attempt in this period or, indeed, later to associate existing archaeological complexes with tribes or peoples known from written sources, even if later sagas and skaldic poetry provide plenty of names for such groups. Thus, there ought to be more to the absence of cultures in Norwegian scholarly discourse about the Iron Age than fear of another Reich. I believe that the reason is deeply embedded in those nationalist sentiments which permeated Norwegian culture and scholarship before and after Norway got its independence from Sweden in 1905, and which implicitly downplayed regional differences by celebrating them as different kinds of expressions of one and the same national culture. After all, it was Anton Brøgger, the

367 Regionality in Iron Age Norway. A Study of Archaeological Cultures 367 Fig. 2. Bjørn Myhre s model of political organization in the Migration period (Myhre 1987)

368 368 Frans-Arne STYLEGAR other nestor besides Shetelig within Norwegian archaeology in the inter-war years, who echoed Kossina most clearly when he argued that archaeology was first and foremost among the national sciences. And working within an eco-functionalist framework, it was Brøgger who used his science for nationalist purposes and gave ideological assistance to Norway s so-called Arctic imperialism in the 1920s and 1930s, including playing a central part in the conflict between Denmark and Norway over East Greenland (Myhre 1994). This national framework has its obvious limitations when dealing with prehistory and early history, both on a regional level and on an international level, in the sense that it makes it more difficult to study regional traditions in their own right and their interrelationship with other regions which often crosses modern state borders. This problem is not a particular Norwegian phenomenon, either. Just a couple of brief examples will show this. The large Roman and Migration period cremation cemetery in Greby in northern Bohuslän, present-day Sweden, has regularly been discussed in a Swedish context, and it is often pointed out that this cemetery is extraordinary in western Sweden, primarily because of its size (Gerdin 1994). But the fact is that the Greby cemetery is normal when compared to cemeteries in Vestfold and Østfold in southeast Norway, like Hunn (Resi 1986), Store-Dal (Petersen 1916), and Gunnarstorp (Wangen 2009) and the Norwegian border is located only 30 km further north. The national border at Svinesund was a result of the Roskilde peace in 1658; until this time, Bohuslän belonged to the kingdom of Norway. Surely, the large cemeteries in Østfold ought to be a natural comparison for the Greby cemetery? Or one could mention the obvious, but unfortunately little-studied parallels between ritual traditions in north Jutland and the southernmost part of Norway in the same period. Typical traits in north Jutland like inhumation burials in large stone cists and wooden, tent-like burial structures are found in south Norway, as well (Stylegar, in press). Again, north Jutland is in most cases being compared to other areas in present-day Denmark, while the basis for comparisons regarding the Norwegian south, remains other areas in Norway (but see: Müller 1912). And this is so even if the links between these two areas across the Skagerrak straitare strong and long-standing, and may well reach back into prehistory (Marstrander 1950). Recently, the study of different ritual traditions in their local contexts is again in vogue. Fredrik Svanberghas provided an interesting analysis of small regions in southern Sweden in the late Iron Age (Svanberg 2003). He was able to identify regional mortuary customs, for instance in the district of Finnveden, where people during the period c had used a special cremation ritual with characteristic mound cemeteries and with specific artefacts deposited in the graves. This burial custom contrasted with those of their neighbours, indicating that this people, during this period, and identified by a specific and unique name, also had their own rituals and customs. Several other more recent studies, often from a Late Iron Age perspective, have stressed this heterogeneity, and focused on small regional identities (Mokkelbost 2007; Urbanzcyk 2003). While such works represent a breath of fresh air, Svanberg s is an exception in the sense that he tries to combine archaeological and written sources. It is also the case that this focus on local traditions may risk missing or downplaying the common denominators running through many of these differences. In other words, there are both small and big ritual traditions. Judging from written sources, it can be argued that large areas, each with their common culture, existed in Scandinavia in the Late Iron Age (see: McLeod 2008, who bases himself on Othere s and Wulfstan s itineraries in the Old English Orosius). But these common cultures, if one can use such a term in this context, did not necessarily follow modern political boundaries (see also Regionalitef 2008). Ritual traditions, great and small What about the time period before the Late Iron Age, then? The place-name researcher Stefan Brink points to the names of tribes and peoples in Scandinavia mentioned by Jordanes and other writers in antiquity. According to him, Scandinavia before the emergence of the kingdoms of Sweden, Norway, and Denmark was characterized by several settlement districts, bygder, or land, which seem to have an older background in ancient folk groups probably existing already in the early Iron Age (fig. 3 4). He quotes studies like Svanberg s approvingly, and urges archaeologists to go further along this route, introducing such neutral examples in the debate regarding the possibility of using archaeological evidence for identifying cultures which bound groups of people together, while avoiding the usual discursive straitjacket furnished by repeated references to the methodological excesses of Nazi, or even Soviet, ethnic archaeology (Brink 2008: ; cf.: Peregrinatio

369 Regionality in Iron Age Norway. A Study of Archaeological Cultures 369 Fig. 3. Iron Age peoples in Scandinavia (Brink 2008) 1992, for a rare excursion into similar territory from a Norwegian point of view). In the following, I will discuss some different aspects of burial customs in Roman and Migration period Norway, focusing on traits with a regional distribution, and with a side-view to the possibility of identifying tribes or peoples known from written sources. There are of course also other, primarily elite traditions cutting across all such boundaries. Weapon burials in bronze cauldrons

370 370 Frans-Arne STYLEGAR Fig. 4. Iron Age provinces in Scandinavia (Brink 2008) is one (fig. 5), chamber graves another (Stylegar 2011; in press). As in Svanberg s Late Iron Age case study from Sweden, there is a whole number of small burial traditions in Early Iron Age Norway, some of them very local (Næss 1996: ; Kristoffersen; Østigaard 2006). In the Trøndelag area, for instance, as in many parts of Norway, most old farms have their own Iron Age cemetery. In the bygd of Oppdal, however, we meet a different pattern. Here we have a handful of very large cemeteries that were used by many farms together, each farm using a designated part of the cemetery (Farbregd 1980). The cemetery at Vang in Oppdal is Norway s largest prehistoric cemetery. Oppdal is a relatively isolated mountain district, and thus perhaps an obvious place to search for local, indigenous rites. The same goes for islands, of course, as evidenced for example by the very different burial traditions met with in Gotland, Öland and Bornholm in the Baltic Sea. From a slightly later period (AD ) than ours we

371 Regionality in Iron Age Norway. A Study of Archaeological Cultures Fig. 5. Roman period weapon burial in bronze cauldron at Hunn, Østfold (photo: Museum of Cultural history, Oslo) 371 have a couple of cemeteries in the island of Jøa in Namdalenwhere people practiced a special rite not found in adjacent areas. The people of Jøa buried their dead in a seated position in pits covered only by low stone settings (Mokkelbost 2007). In the interior of east Norway, Lake Mjøsa separates the ancient districts of Hadeland and Hedemarken, the latter once home of the heinir, the former home of the haðar (Andersson 1999a d). In our period, as well as later periods, cremation is the dominant rite in both areas. But there are still some interesting differences between the two areas. On the western side of Lake Mjøsawe find the practice of urned burials in small stone cists; sometimes the urns are Roman bronze cauldrons, but as often they are ceramic vessels. In Hedemarken, on the other hand, we do not find similar stone cists, and pottery vessels are either used as containers for cremated bones or they are part of the furnishings in regular cremation layers (Martens 1969: 64). The prevailing cremation rite in Hadeland as well as in the interior of east Norway as such, is often interpreted as a conservative trait. This period is usually called the Migration period, but nowhere in Norway is this name as unfitting as in here, where the culture during these centuries is completely dominated by old traditions, Sigurd Grieg writes about Hadeland (Grieg 1926: 97). In Jæren, southwest Norway, we meet a particular tradition throughout the Roman and Migration periods. In a region which is otherwise distinctly biritual, there are several large cemeteries situated on the beachfront of Jæren where inhumation is the only rite found (Lillehammer 1996). Inhumations appear in these beach cemeteries in the Early Roman period already, unlike in any other district in south Norway. Thus, these cemeteries have sometimes been interpreted as belonging to an intruding population (Mortensen 1992), even to the rygir mentioned by Jordanes. But the most convincing case for locating tribal burial traditions comes from north Norway. The coastal strip of the vast tracts of land north of Trøndelag was called Hálogaland the land of the háleygir (Andersson 1999e). The burial rites in this area are relatively homogenous, and characteristically different from those in all other areas in present-day Norway. To start with, cremations are practically unknown here, while it is inhumations which make up a minority in all other regions, including the adjacent Trøndelag (Sjøvold 1962: 145). Flat graves are almost unknown, too most Iron Age graves are covered by relatively small mounds. The situation of the cemeteries is also very characteristic: many of them are situated on small islands or islets at some distance from the settlements (Evensen 2003). To the north and east of Hálogaland was Finnmork, where the finnar or Saami lived. This is far beyond the Arctic Circle, where no agriculture was possible, and the Saami in our period were hunters and fishers. The burial rites of Finnmork are as characteristic as the ones in Hálogaland. The normal form of burial are the so-called scree graves, where the dead was interred in a scree, under a boulder, in a cave or in a small opening in the fjell (fig. 6). The dead body was sewn into birch bark. Sometimes, the remains of a pulk or toboggan are found, as well. The oldest dated scree graves come from the Varanger peninsula, where Mortensnes with its close to 300 graves is the largest cemetery in north Norway (Schanche 2000).

372 372 Frans-Arne STYLEGAR Fig. 6. Samii scree grave, Vadsø, Finnmark (Myrvoll 2010) Some ritual traditions covers substantially larger areas than settlement districts, and it is difficult to associate these with tribal areas of the type discussed by Brink, which for the most are relatively small, naturally limited districts. This could be argued for the long Hálogaland coast, too, but it is clear for all to see in cases like the large stone cists in western Norway (fig. 7 9). This characteristic type of grave was noticed already by Oluf Rygh in He described them as carefully built chambers of considerable proportions in most cases longer than an adult person but as a rule relatively low and narrow The building material is stone; the sides are either made from smaller boulders or from more or less flat stones, in most cases flagstones standing upright. The roof is always composed of one or more huge stones, usually flagstones (Rygh 1869: 62). The inner construction of these graves is very homogenous. The floor in many cases has a bottom layer of small pebbles, with birch bark and then an animal skin, often a bear fell, followed by blankets on top. Extended burial is common, and the body is fully clothed and equipped with personal ornaments, jewellery etc. In many men s graves, weapons are found along the body, with the shield in most cases covering the deceased s legs. Urns, either locally produced ceramic ones or imported vessels made from copper alloy or glass, are always in one of the gable ends, in most cases the foot-end; this part of the cist is in some cases organized as a separate chamber (Schetelig 1912: 110 f). The large stone cists appear early in the Late Roman period, but the distribution peaks in the Migration period. While they are not completely unknown in eastern Norway and the Trøndelag area, they are to a large degree a regional phenomenon in western Norway. The vast majority of the finds are from the western region comprised of the modern counties Vest-Agder, Rogaland, Hordaland, and Sogn og Fjordane, and the largest of these graves are all from this area. The same area is distinguished also by its material culture. The border between the eastern and the western Migration period complex is the same as the eastern border of the distribution area of large stone cists. And some artefact types, like bucket-shaped pots, lathe-turned spindle whorls made from stone, and bone implements like spoons and meat-cleavers are particular to the region (Brøgger 1909; Engevik 2008; Stylegar 2013). Would it really be out of order to call this a culture in the archaeological sense? Regional traditions and the question of cultures It gets more intriguing if we compare this western area to other, larger areas. Taking our lead from the fact that just this area of western Norway was a judicial as well as a political entity in the Late Iron Age the Gulatingslag with its periodic regional assembly of leading men in Gulen in Sogn (Helle 2001), it may be a good idea to look at similar entities in a later period than ours and ask whether these also had their own ritual traditions in the Roman and Migration periods, thus setting them off from other areas of roughly the same size and dignity. According to the historian Sverre Steen, the political situation in present-day Norway in the Early Medieval period was characterized by a number of different political entities: Vík, the Oslofjord area from Cape Lindesnes and eastwards, was controlled by Danish kings; our area in the coastal districts of western Norway was a kingdom in its own right, the Norðmanna land of Othere and other sources, sometimes in alliance with Thrændalog; and one or more small kingdoms in the mountain region of Upplond, with uncertain and somewhat blurry borders with Víkin the south, and with obvious connections to the east, i. e. to Sweden (Steen 1942: 96). We might also add Hálogaland to this list, and Finnmork (fig. 10). While Steen s choice of terminology ( kingdoms ) is perhaps somewhat anachronistic, and while we would rather speak of tribal confederacies or even archaic states, these large regions are of interest to us. There is no good reason why in

373 Regionality in Iron Age Norway. A Study of Archaeological Cultures 373 Fig. 7. Large stone cists have a marked western distribution (illustration: the author)

374 374 Frans-Arne STYLEGAR Fig. 8. Large stone cist under excavation at Lista, Vest-Agder (photo: Museum of Cultural history, Oslo) principle these areas should not correspond to cultural areas, as well. But was this the case already in the Roman and Migration periods? As should be evident from the above, this is pretty clear when it comes to what we may refer to as Norðmanna land, and also when it comes to Hálogaland. There is little doubt that archaeologically speaking, eastern Norway is different from these other areas. From the fourth century onwards, eastern graves become more and more simple, as seen from the large cemeteries at Hunn and Store- Dal. There are a small number of richly furnished inhumations, but such finds only serve to put the dominant type of burial in relief: A simple crema-

375 Regionality in Iron Age Norway. A Study of Archaeological Cultures 375 Fig. 9. Large stone cist from Gloppen, Sogn og Fjordane (Schetelig 1912) tion rite, mostly cremation layers with some pottery (fig 11). The few inhumations there are, are often in large pits, very different from the stone cists in the west. Urn burials in or below mounds are pretty rare. In western Norway, however, urn burials in or below mounds is a significant trait in the Migration period, along with inhumation in large cists. The border in this respect between the two areas seems relatively clear-defined, and is concurrent with Cape Lindesnes in present-day Vest- Agder; to illustrate the point, it might be pointed out that the Migration period Stoveland cemetery slightly to the east of Cape Lindesnes has cremation layers only and no urn burials (Rygh 1879), while in the settlement district of Lista immediately to the west of the Cape, urn burials dominate (Grieg 1938). At the large cemetery in Spangereid, situated at the Cape itself, however, both forms are known (Rygh 1880). Furthermore, in the Oslofjord area the cremated bones are usually deposited together with ashes and other remains from the funeral pyre, while cremation burials in western Norway in the majority of cases have cleansed bones (Schetelig 1913). Again in the Oslofjord area, and unlike in western Norway, large cemeteries are relatively common (Løken 1974). Substantial cemeteries with urn burials without any visible marking above ground, or within different types of stone settings, are also relatively common in the Oslofjord area (Johansen 1955; Munch 1965). Cremation is the dominating rite in the inland region of Upplond, and cremation pits is the most typical type of burial there (Martens 1969: 36; Herteig 1955: 74) unlike the Oslofjord area, where cremation pits occur more rarely and are replaced by cremation layers in the Roman period (Hougen 1932: 23), and unlike western Norway, where they barely register. Apart from this, the inland of eastern Norway is characterized by its many weapon graves dating to the Late Roman period (fig. 12). Finally, Trøndelag has a mix of inhumation and cremation burials, but with a clear majority of cremations (Johansen 2012). These are typically urn burials, with only a few cremation pits (Pe tersen 1923). A particular type of cremations in small boxes or stone cists, without an urn, is also relatively common in Trøndelag, but practically unknown in other parts of Norway (Marstrander 1983). The cremation layers which become increasingly common in southeast Norway during the Late Roman and Migration periods, are rarely found in Trøndelag. These six larger ritual traditions in the Oslofjord area, in the mountainous Uplands, on the coast of western Norway, in Trøndelag, in

376 376 Frans-Arne STYLEGAR Fig. 10. Major landscapes of the Late Iron Age, according to written sources (illustration: the author)

377 Regionality in Iron Age Norway. A Study of Archaeological Cultures 377 Fig. 11. Mound with cremation layer, Ula-Glemmen, Østfold (Vibe-Müller 1987)

378 378 Frans-Arne STYLEGAR Fig. 12. Late Roman period weapon burials: red cremations; black inhumations (illustration: the author)

379 Regionality in Iron Age Norway. A Study of Archaeological Cultures 379 Fig. 13. Major ritual regions in the Roman and Migration periods (illustration: the author)

380 380 Frans-Arne STYLEGAR Hálogaland and in Finnmork it seems reasonable to interpret as different material culture complexes cultures, if you like (fig. 13). I have focused my attention on burial rites, but there are also other, characteristic differences, only some of which I have mentioned here, like the distribution of bucket-shaped pots or the dividing line at Cape Lindesnes between western and eastern artefact complexes. Interestingly, some of the regional differences discussed here seem to become clearer and more pronounced during the period under discussion (AD 1 550). Typical examples are the stone cists in western Norway and the cremation layers in the Oslofjord area, both of which are very much Migration period phenomena. These and the four others are the same areas met in the Late Iron Age and Early Medieval written sources in the shape of tribal confederacies or proto-kingdoms which by the end of the Viking Age came together in the medieval kingdom of Norway. It is often believed that these political entities are of Viking Age date the archaeological sources, however, suggest that they might have a longer history, at least as material culture and burial rites go. Conclusion Regionality, in the specific archaeological sense of the spatial distribution of differences in material culture, has not been a central concern for Iron Age specialists in Norway for several decades. But with its varied topography, its mountains, fjords, rivers and lakes, the country is an obvious place to look for such differences. In the present paper, I have turned my attention to burial customs. As one would expect, there is a myriad of local and regional rites to be found, and some traits might occur all over Norway, and indeed over the whole Germanic area. However, some of these ritual traditions might be limited to a single farm, or even to a single family in some cases, while others are probably associated with distinct tribes or other larger groups. But then there are some ritual traditions which cover substantially larger areas than nature-given settlement districts, but still have a rather limited distribution. I have tried to show that these larger areas concur with the major regions met with in later, written sources dealing with the origins of the Medieval kingdom of Norway. Bibliography Andersson T. Haðar // RGA. 1999a. 13. Andersson T. Hadeland. Namenkundliches // RGA. 1999b. 13. Andersson T. Hedemark. Namenkundliches // RGA. 1999c. 14. Andersson T. Heinir // RGA. 1999d. 14. Andersson T. Háleygir // RGA. 1999e. 13. Brink S. People and Land in Early Scandinavia // Franks, Northmen, and Slavs: Identities and State Formation in Early Medieval Europe. Brepols, Brøgger A. W. Benskeer fra ældre jernalder // Årbok Stavanger Museum Engevik A. Bucket-Shaped Pots. Style, chronology and regional diversity in Norway in the Late Roman and Migration periods. Oxford, Evensen B. Gravholmer. Gravlokalisering og samfunn i jernalder i Vesterålen (Master s Thesis). Tromsø, Farbregd O. Gravfeltet på Vang // Bøgda vår Oppdal, Gerdin A.-L. Svärd och säd från bronsåldern samt mycket annat om nya spännade fynd i Tanum. Kungsbacka, Gjessing G. Opphavet til håløygjarlenes rike // Håløygminne. 10 / 2. Harstad, Gjessing G. Mer om ophavet til håløygjarlenes rike // Håløygminne. 11 / 1. Harstad, Gjessing H. Vest-Agder i forhistorisk tid // Norske Bygder. II. Vest-Agder. I. Bergen, Grieg S. Hadelands eldste bosetningshistorie. Oslo, Grieg S. Listas jernalder. Oslo, Helle K. Gulatinget og Gulatingslova. Leikanger, Hougen B. Oldtidsminne, funn og bygd. Introduction // Østfolds Oldtidsminner. Oslo, 1932.

381 Regionality in Iron Age Norway. A Study of Archaeological Cultures 381 Johansen E. Ny datering av branngraver under flat mark. Gravskikken som kilde til sosial historie // Årbok Universitetets Oldsaksamling Oslo, Johansen T. Trekk av eldre jernalders dødekult i Trøndelag // Graver i veien. Arkeologiske undersøkelser E6 Steinkjer. Trondheim, Klejn L. S. Kossina im Abstand von vierzig Jahren // Jahreschrift mitteldeutschen Vorgeschichte Klejn L. S. Gustaf Kossinna: // Encyclopedia of Archaeology. The Great Archaeologists. 1. Santa Barbara, Kristoffersen S., Østigaard T. Dødsmyter regissering av ritualer og variasjon i likbehandling i folkevandringstid // Lik og ulik. Tilnærminger til variasjon i gravskikk. Bergen, Lillehammer G. Død og grav. Gravskikk på Kvassheimfeltet, Hå i Rogaland, SV Norge. Stavanger, Løken T. Gravminner i Østfold og Vestfold. Et forsøk på en typologisk-kronologisk analyse og en religionshistorisk tolkning / Master s Thesis. Oslo, Marstrander L. Inntrøndelag i romertid. Gravfunn og bosetning. Trondheim, Marstrander S. Jylland Lista // Viking Martens I. Gravfeltet på By i Løten, Hedmark // Årbok Universitetets Oldsaksamling Oslo, McLeod S. Know Thine Enemy: Scandinavian Identity in the Viking Age // Vikings and their Enemies. Melbourne, Mokkelbost M. Sittegravbyggerne i Sandvika. En arkeologisk analyse av en lokal tradisjon i yngre jernalder på øya Jøa i Ytre Namdal / Master s Thesis. Trondheim, Müller S. Vendsyssel-Studier III // Aarbøger for nordisk Oldkyndighed og Historie. København, Munch J. S. Borg og bygd. Studier i Telemarks eldre jernalder // Årbok Universitetets Oldsaksamling Oslo, Myhre B. Chieftains graves and chieftain territories in south Norway in the Migration Period // Studien zur Sachsenforschung Myhre L. N. Arkeologi og politikk. En arkeo-politisk analyse av faghistoria i tida / Master s Thesis. Oslo, Myrvoll E. R. Kulturminner i «Askeladden» og kategorienes vold // Viking Næss J.-R. Undersøkelser i jernalderens gravskikk på Voss. Stavanger, Peregrinatio Gothica. III. Oslo, Petersen J. Gravplassen fra Store-Dal i Skjeberg. Kristiania, Petersen Th. Meldalsfunnene. En gravplass under flat mark fra ældre jernalder paa Vahaugen i Meldalen. Kristiania, Petersen Th. Problemer i det nordenfjeldske Norges bosetningshistorie // Det Kongelige Norske Videnskabers selskabs forhandlinger. 3. Trondheim, Regionalitef i Danmakk i vikingetid og middelalder. Højbjerg, Resi H. G. Gravplassen Hunn i Østfold. Oslo, Rygh O. Om den ældre jernalder i Norge // Aarbøger for nordisk Oldkyndighed og Historie. København, Rygh O. Undersøgelser paa en Gravplads fra ældre Jernalder i Holme Sogn ved Mandal // Fortidsminneforeningens aarsberetning Kristiania, Rygh O. Gravundersøgelser paa Spangereid // Fortidsminneforeningens aarsberetning Kristiania, Schanche A. Graver i ur og berg. Samisk gravskikk og religion fra forhistorisk til nyere tid. Karasjok, Schetelig H. Vestlandske graver fra jernalderen. Bergen, Schetelig H. Den førromerske jernalder i Norge // Oldtiden. III. Kristiania, Shetelig H. Et norsk folkevandringsrike // Aarbøger for nordisk oldkyndighed og historie Copenhagen, Sjøvold T. The Iron Age Settlement of Arctic Norway. 1. Oslo, Slomann W. Folkevandringstiden i Norge. Spredte trekk og enkelte problemer // Årbok Stavanger Museum Stavanger, Steen S. Ferd og fest. Reiseliv i norsk sagatid og middelalder. Oslo, 1942.

382 382 Frans-Arne STYLEGAR Stylegar F.-A. Weapon Graves in Roman and Migration Period Norway (AD 1 550) // Inter Ambo Maria: contacts between Scandinavia and Crimea in the Roman period. Kristiansand; Simferopol, Stylegar F-A. Dreide spinnehjul av bergart fra folkevandringstid // Nicolay Stylegar F.-A. Chamber Graves and Other Oversized Graves in Roman and Migration Period Norway // Kammergräber in Barbaricum. Neumünster (in press). Svanberg F. Decolonizing the Viking Age. Vol. I. Death Rituals in South-East Scandinavia AD Stockholm, Undset I Fra Norges ældre Jernalder // Aarbøger for nordisk Oldkyndighed og Historie. København, Urbanczyk P. Breaking the Monolith: multi-cultural roots of the North Atlantic settlers // Vinland Revisited: The Norse World at the Turn of the First Millennium. St. John s, Vibe-Müller K. Gravfeltene på Ula, Glemmen, Østfold: Keltisk jernalder, romertid og folkevandringstid. Oslo, Wangen V. Gravfeltet på Gunnarstorp i Sarpsborg, Østfold. Et monument over dødsriter og kulturutøvelse i yngre bronsealder og eldste jernalder. Oslo, Франс-Арне СТЮЛЕГАР Регионализм в Норвегии в железном веке. Исследование археологических культур Резюме Проблема регионализма, который в специфически археологическом смысле означает пространственное распространение различий в материальной культуре, в течение нескольких десятилетий не был главным вопросом, обсуждавшимся норвежскими специалистами по железному веку. Тем не менее, данная страна благодаря своей разнообразной топографии, включающей горы, фьорды, реки и озёра, является тем местом, на материалах которого и следует изучать проблему регионализма. Автор данной статьи обращается к погребальным обрядам. Как и следовало ожидать, здесь можно обнаружить бесчисленное множество локальных и региональных обрядов, некоторые из которых встречаются на всей территории Норвегии и даже по всей области расселения германцев. Однако часть обрядовых традиций фиксируется лишь в пределах одной сельской усадьбы, а в некоторых случаях даже одной семьи, тогда как другие связаны с отдельными племенами или другими крупными объединениями. В то же время, имеются некоторые обрядовые традиции, встречающиеся на существенно больших территориях, чем зоны поселений, ограниченных природными условиями, но тем не менее, достаточно слабо распространённые. Как представляется, имеет смысл интерпретировать эти шесть крупных обрядовых традиций в районе Осло-фьорда, на Оппланнском нагорье, на побережье западной Норвегии, в Трёнделаге, Халогаланде и Финнмарке как разные материальные комплексы, культуры. По мнению автора, эти крупные обрядовые традиции соответствуют крупным регионам, о которых пишут более поздние письменные источники в связи с происхождением средневекового норвежского королевства.

383 383 Jaroslav TEJRAL The Connections between the Region North to the Danube and Northem Europe. Some Aspects of Ethnic and Social Identity of the Fifth Century Elites The main goal of the lecture is to point out to certain kinds of archaeological evidence with connections to Scandinavia and the Baltic and to make an attempt to identify the bearers of the archaeological vestige in the Middle Danube area in the fifth century, at the time before the domination of Lombards in this region. Archaeological research of the Great Migration Period (fifth sixth century) on the Middle Danube is repeatedly confronted with the question to what extent the assessments of archaeology can be used to draw reliable historical conclusions. Particularly during the fifth century, which was characterised by strong mobility on the part of individuals and smaller groups of population as well as of larger tribal communities, discussion focuses on the problem of whether the individual ethnic groups can be identified on the basis of their material culture, thus providing an archaeological proof of the migratory movements. The issue of migration and settlement changes is of special significance when assessing archaeological developments to the north of the Danube, in what is now the Czech Republic, northern Austria and south-western Slovakia. In resolving the question to which extent archaeology can identify ethnic entities or migrations at all, the material from this region occupies a special position. The archaeological record here dated to the fifth and sixth century, which manifests predominantly in cemeteries, is not uniform, but rather culturally diverse. The earlier group of archaeological evidence shows very similar features as developed throughout the entire Middle Danube region during the fifth century; the second, marked by the prevalence of Elbian-Germanic handmade pottery and the frequent presence of Merovingian weapons, has been associated with the Lombards or related ethnic communities and dated in the sixth century. The actual relationship of the older group of burial grounds to the later is of crucial importance for the much-debated issue of the settlement continuity in the region, and this problem is more complex than previously believed (Bemmann 2008: ; Quast 2010b). The first, earlier group of archaeological remains, cultural configuration and burial rites of which were markedly different from that of the previous Late Roman period, began to appear by the second third of the 5th century in the content of isolated inhumation graves, family burial sites and larger cemeteries, as well as in the archaeological record from certain number of settlements. A series of skeleton burial sites in the region of Moravia is known as at Novy Šaldorf (72 graves), Velatice (34 graves), Vyškov (more than 30 graves), Šaratice (more than 35 graves), Strachotin (21 graves), Sokolnice etc. (Staňa 1956; Tejral 1982: 66 75, , ; Tejral 2005: , , Abb. 1 8; Čižmář, Geislerová, Rakovský 1985). Smaller burial sites with fewer (about 12) graves was completely unearthed in Břeclav-Líbivá (Macháček, Klanicová 1997) and Vrchoslavice, while analogous sites have also been found in northern lower Austria, such as at Mitterhof and Schletz (Heinrich 1990; Lippert 1968; Windl 1988; Windl 1996; Windl 1997). The latest finds suggest that the large cemeteries (Nový Šaldorf, Strachotín, Šaratice I, Velatice, Vyškov) and also some smaller ones (Schletz) continued to function until the late fifth, possibly early sixth century. Some separate graves or groups of female and male inhumations of the same cultural origin also show dates, which may go as late as ca 500 (Tejral 2005: , Abb. 5; Čižmář, Tejral 2002). The settlements in the region are mostly small homesteads with recessed dwellings with supporting posts in the corners, characterised by plentiful fast-wheel pottery: either coarse utility ware in the

384 384 Jaroslav TEJRAL style of Late Antique pottery, or fine Danubian type pottery often adorned with bold polished decoration (Tejral 1982: 75 81; Tejral 1990: 34 42; Tejral 2005: ). On at least five sites (Mušov, Peigarten, Rajhradice, Cholina), archaeologists found the potters kilns used to make the ceramics. Characteristic types of combs and brooches, in conjunction with other finds, have allowed one to date these settlements to the middle and second half of the fifth century (Tejral 1982: 206, Abb. 79; Čižmář 1999; Kern 2000; Přichystal, Vachůtová 2007). The attempt to resolve the question of ethnical ascription of the north-danubian archaeological record and who was the driving force behind the historical events in the that Danube region during the fifth century encounters major difficulties. The striking uniformity of the archaeological record from inhumations in the broad zone from the Middle Danube region nearly to the Balkans during the fifth century confirms the opinion that there existed a relatively homogeneous cultural circle in this area, which developed parallel to the proto- and early stage of Merovingian culture in the West and, despite mutual influences and a number of common features, retained its own individual character. There are also evident genetic links between this and the later material culture of so-called Gepidian large cemeteries in the Great Hungarian plain (Csallány 1961; Bóna 1976; Bóna, Nagy 2002; Cseh et al. 2005) and even with Eastern Germanic small burial grounds in Italy or in the Balkans (for example Bierbrauer 2007: ; Miletić 1975). The phenomenon of the establishment of analogous cemeteries and settlements in such a vast area may be down to a number of factors, the most relevant of which would be the definitive formation of socio-political structures within the individual barbarian tribal groups and barbarian kingdoms, which were the result of more complex overlaying and assimilation processes during the fifth century. Although the textual sources refer to these formations with specific tribal names, the ethnic relations within them might have been more intricate. The names of various tribes, considered as an ethnic designation, may instead relate to a social group. Since the archaeological record corresponding to the so-called Danubian Eastern Germanic cultural circle reflects the specifics of life and fashion of numerous peoples which were settling in the Middle and Lower Danubian area and may in great part be seen as an expression of regionality rather as ethnicity. Nevertheless some single and specific features of the dressing customs, burial gifts and other items which manifest an awareness of the difference of other contemporaries or the peculiarities of burial rites may, according to some authors, indicate the bearers as belonging to a special social and in this juncture possibly also ethnic group. Besides the phenomena which reflect in both archaeological material and anthropological findings a strong influence of the equestrian nomadic heritage, it is the sphere of relics whose nature indisputably refers to the traditions of earlier Eastern European cultures. One characteristic feature of the extraneous ethnic element, which clearly also has a social dimension, are the single graves or small cemeteries with graves containing prestigious, socially significant parts of dress accesories completed with further symptomatic elements of grave equipment. This was the case, for example, in the female graves with their large Eastern Germanic brooches made of sheet silver and other spectacular parts of clothing (fig. 1, 2), such as in Smolín grave no. 32 or Laa an der Thaya grave no. 1 (Tejral 1973; Tejral 2007b: 85 90, Abb. 16, 18; Beninger 1929). The single and isolated groups of warrior graves, like those in Vienna Leopoldau, Sigmundsherberg (Beninger 1936; Friesinger 1984; Szameit 1984), Velatice, Držovice (Tejral 1982: ; Čižmář, Tejral 2002) are also indicative. Their equipment, mostly comprising eastern weapons, typical Asiatic swords or narrow nomadic seaxes often accompanied by other artefacts of eastern provenance (e. g. crescent-shaped earrings) is characteistic of this form of grave (fig. 3, 4), which is widespread in Central and Eastern Europe from Przemęczany in southern Poland to Košarevo in Bulgaria (Godłowski 1995: , Abb. 13; Daskalov 1998). In some larger cemeteries as in Velatice were analogous nomadic weapons found together with skeletons showing cranial deformations (fig ). As a supra-regional phenomenon could be linked that burials with the social stratum of the elite warriors of nomadic, and also of Eastern Germanic origin. A special case means the rich princely grave at Blučina-Cezavy with hilt-gilded spatha and other golden items (Tihelka 1963). The symbolic gift of a nomadic reflex-bow and trilobate arrowheads (fig. 5) manifest the strong equestrian nomadic influence on the burial equipment of local authorities. Cranial deformations often associated with this archaeological record and a certain proportion of mongolid and asiatic anthropological markers among the skeletal material indicate kinship be-

385 The Connections between the Region North to the Danube and Northem Europe. Some Aspects of Ethnic and Social Identity of the Fifth Century Elites 385 Fig. 1. Female burial no. 32 from Smolín (Moravia, Czech Republic), showing typical brooches of sheet silver, golden earrings, and other silver dress accessories tween the population buried in the cemeteries north of the Danube and the groups settled in the rest of the Danubian area, which share the same features (Anke 1998 / 1: ; Tejral 2005: ; Straub 2011: ; Tobias, Witschke-Schrotta, Binder 2010: , ). Nevertheless, the problem is still open, thereby calling for new methodical processing. Even though it cannot be considered a clearly defined and discrete group, a small assemblage of finds which exhibit distinct relations to the European north occupies a special position. The evidence of cultural contacts between the Middle Danube region and regions in the north of Europe can, however, be followed up within a more general and chronologically as well as territorially broader scope since at least the first half of the 5th century already. As an evidence of far-reaching influences of the Danubian equestrian nomadic realm in Scandinavia have usually been named the hoards of silver and bronze objects found in bogs at Sösdala, Fulltofta and Vännebo in Southern Sweden. These hoards, which are being considered a Nordic parallel to Hunnic sacrificial finds of the Pannonhalma type, included metal components of horse harness, saddles etc. (Fabesch 1991). Impulses from the Danubian region played here also most probably an important role in the emergence of Nordic stamped artefacts in the Sösdala style. Besides a rein guide with bird-head ends (Bemmann 2006: 228; 2007: 178, Abb. S. 179), the Danubian influence is also visible in several types of chip-carved artefacts (Näsman 1984b: 65 67), among other thinks also brooches with semi-circular head-plate (Bemmann 2008: 170, Abb ). In this early period, however, typical Scandinavian objects are found only seldom in more southerly regions. It was mainly in the last time that in this context the Poskaer-type dividers of sword scabbard (Menghin 1983: 139, 304; Miks 2007: , Abb. 105; Bemmann 2006: ) were found for example at the Polish site of Spiczyn (Luckiewicz 2008: 220, ryc. 2 3; Quast 2010a: , Abb ) and in the burial ground at Epöl on the territory of what was once the province Pannonia (fig ), which indicates relations of the southern Scandinavian area to more southerly situated continental regions (Bemmann 2006: , Abb ). The latest find of a single example from Bohušov in Czech Silesia (unpublished find, kindly shown to me by Petr Kubín,

386 386 Jaroslav TEJRAL Fig. 2. Brooches of sheet silver and selected objects from the female grave no. 1 at Laa an der Thaya (northern Lower Austria) Regional Museum Mikulov) proves that the communication between north and the region on the middle Danube was maintained even through the Moravian Gate. More attention in this regard deserve the finds from the territory north of the Middle Danube, above all from Moravia. Important is mainly a small burial ground at Vrchoslavice in Central Moravia, particularly grave no. 12. It yielded a Nordic cruciform brooch with rhombic foot (fig. 7. 5), which finds parallels at the Scandinavian site of Uppåkra and in other places (Hårdh 2003: 46 51, fig. 1. 1). This brooch, however, is not isolated north of the Danube and some kind of analogy (fig. 7. 3) was found at the hilltop site of Bojná II in Southern Slovakia (Pieta 2006: , obr. 3. 3). This site also yielded other finds which are datable to around the mid of the 5th century or to the second half of the same century, among others two nomadic narrow seaxes, maybe from disturbed graves (Turčan 2003: , obr. 2). Grave 12 from Vrchoslavice yielded also some other important finds. Among them was a second brooch with relation to the north, namely a Pomeranian variant of the Wiesbaden-type brooch (fig. 7. 6), which finds close analogies for example in the find no. 300 from the settlement Dębczyno, among brooches in the treasure hoard at Trzebiatów etc., but also in the find from Foss, Norway (Godłowski 1979: Abb. 13, 16b; Werner 1981: , , Taf , Beil. 2, 21, 24). One of the three glass vessels which were found in the grave, namely a high footed beaker with applied glass threads (fig. 7. 4), a variant of Eketorp glass 8 according to Näsmann or of Snartemo type II, also exhibits relations to the north (Näsmann 1984a: 72; Stawiarska 2000: 154, ryc. 3. 2) and its nearest parallel can be found in Barshaldershed on Gotland (Stjernquist : 159, fig. 16). Further examples of Danubian glass vessels of similar type (Kapušany, Ghenci, Prša, Nitra, Oberleiserberg) and their pedestalled sister variant of the Barabás-

387 The Connections between the Region North to the Danube and Northem Europe. Some Aspects of Ethnic and Social Identity of the Fifth Century Elites 387 Fig. 3. Examples of Danubian warrior graves with eastern weapons. 1 6 grave no. 3 from Wien-Leopoldau; 7 13 grave no. 2 from Držovice Kosino type (Tarnamera) form a discreet group which is well-datable to the second third of the fith century (Tejral 2011: 45 46, Abb. 17). A conspicuous find which also implies relations to the north is a bronze pattern of the rectangular scabbard mount of Nydam II type decorated with geometric chip carving in the Nydam style (fig. 7. 1). It is a type documented so far almost exclusively in Scandinavia, particularly in ritual depositions from Ejsböl or Nydam II but also in other North-European graves (Bemmann, Hahne 1994: , ; Miks 2007: , Abb. 94). From the territory outside of Scandinavia comes also a find from a sumptuous burial of a warrior at the Baltic site of Taurapilis (Bliujienė, Steponaitis 2009: 188, fig. 8. 6, pl. IV. 3). Possible production of this type of fittings in the Middle Danube region, however, raises further questions about the nature of mutual Danube-Scandinavian relations. However conspicuous are the similar finds showing clear relations to Scandinavia or to Pomerania and Baltic area respectively, it would undoubtedly be premature to associate them with particular ethnic element. No matter how many questions remain

388 388 Jaroslav TEJRAL Fig. 4. Weapon graves with cranial deformations and other items from the burials in the cemetery at Velatice conjectural, it is likely that there was no clear-cut ethnic situation in the region to the north of the Middle Danube and we can assume that there was a very diverse range of ethnic groups. Besides, nations and tribes referred to in written sources by specific ethnic names do not make up the entire population of a given territory, but rather its leading social stratum. According to textual sources the main power factor in the Middle Danube area there were Heruli, supposed of Scandinavian origin (Ellagård 1987; Taylor 1999) which are mentioned by Julius Honorius between the Quadi et Macomanni already at the beginning of the fifth century (Pohl 1980: 277; Wolfram 1995: 58; Lotter 2003: 103, 130). In any case the Heruli are recorded as having been involved in the battle of Nedao ca. 456 AD (Heather 2008: , and others). They were major players in the Middle Danube region until the beginning of the sixth century. Together with other Danubian tribes they participated in Odoaker s (rex Herulorum) campaigns in Italy. Somewhere here, to the north of the Danubian frontier, led by the famed King Rodulf, clashed with the Lombards by them the Heruli have been routed (on the last, Pohl 2008: 5 6). The historically substantiated Heruli in this case may only have comprised the narrow social

389 The Connections between the Region North to the Danube and Northem Europe. Some Aspects of Ethnic and Social Identity of the Fifth Century Elites 389 Fig. 5. Princely grave from Blučina-Cezavy, selected finds. 1 Hilt-gilded spatha; 2 9 iron arrowheads; silver saddle fittings; symbolic gift of a nomadic reflex-bow

390 390 Jaroslav TEJRAL Fig. 6. The distribution of Poskaer-type dividers of sword scabbard stratum of elites, ruling, dominating a number of groups including perhaps the remnants of the local Suebi and fragments of other tribal communities which were originally from a variety of different places. Otherwise this is also backed up by the results of the latest research, which sees the groups referred to as the Heruli in various places throughout Europe not as a separate tribe, but as factions of elite warriors, as the reports from late antiquity claim (Ellagard 1987; Taylor 1999). The sudden change in the archaeological record reflects the second, later group of burial grounds

391 The Connections between the Region North to the Danube and Northem Europe. Some Aspects of Ethnic and Social Identity of the Fifth Century Elites 391 Fig. 7. Some finds with cultural relations to the European North in the region north to the Danube. 1 bronze pattern of the rectangular scabbard mount of Nydam-Eisbøl type from the hoard at Steinmandel (Buschberg); 2 rectangular scabbard mount from Nydam; 3 cruciform brooch from Bojná II; 4 6 high-footed glass beaker and brooches from grave no. 12 at Vrchoslavice in the region north to the Middle Danube and posed an essential problem, which is hotly debated. The change is manifested by the emergence of a large number of burial sites with grave assemblages, which are, distinctive examples of the Elbian-Germanic or Thuringian cultural pattern and which differ in many respects from the earlier group of graves (Bemmann 2008: ). The largest number of such finds is spread across the area of Lower Austria above the Danube and in southern Moravia. In the recent years, more cemeteries have been thoroughly excavated: 107 graves in Holubice, 56 in Šakvice, 36 in Čejkovice, 23 in Borotice, 120 in Lužice near Hodonin, about 40

392 392 Jaroslav TEJRAL in Mušov and 240 in Kyjov (Tejral 1976; Tejral 2005: , , Abb ; Tejral et al. 2011a). The burial sites in this group feature a striking quantity of diverse handmade pottery, basically bowl-shaped vessels with a decor implying their Elbian origin. The grave goods differ from those of the previous group of burial grounds mostly also in the small finds that survived the looting, which was common. Female graves contain some wonderful silver gilded bow brooches, to be worn on the lower part of the dress, or the cingulum-like pendants and small, predominantly S-shaped plate brooches fastened on the chest. Male graves often contain weapons: swords (ordinary Merovingian spathae), shield fittings, various types of lances and spears or throwing weapons, including some outstanding long spears of Frankish origin, known as ango. Various types of axes are also common. Both male and female graves contain bone or antler combs. Thus, regardless of some of generally Merovingian items, the earlier and the later group of cemeteries differ considerably on account of the grave-goods. Significant suggestions in archaeological record in Moravia and Lower Austria counters the theory of continuous development of the native population and its background. At the turn of the fifth sixth century, there is evidence of a major change in settlement patterns. Most importantly, the burial sites from the second half of the fifth century and the later inhumation cemeteries with handmade Elbian-Germanic pottery, show no developmental connections in the sense of spatial succession. The absolute majority of the Lombard burial sites were founded in other locations than the cemeteries of the previous period. Development discontinuity is evidenced by the sudden and radical decline of the settlements dated to the fifth century: none of them survived into the 6th century. This phenomenon is accompanied by a collapse of the local manufacture of wheel-made ceramics. Doubts are raised by the fact that this metamorphosis about AD 500 only occurred within the limits of southern Moravia and Lower Austria, while in the eastern part of the north Danubian zone, in modern Slovakia, the penetration of the distinctive Elbian-Germanic-Thuringian cultural elements was only marginal (Tejral 2012: ). Though still preliminary results have been obtained by Sr, Pb and O isotope analyses in tooth enamel of the deceased analysed samples from the later, so-called Lombard phase at two key cemeteries of Lužice and Holubice in Moravia have revealed that about a third buried individuals were of foreign origin. Assuming that these cemeteries functioned for three generations, this evidence shows the crucial portion of a foreign element at their foundation (Tejral 2012: 140). The achieved results indicate already now not only a change in cultural habitus at the time at AD 500, but also suggest that the rupture of the hitherto development might have been caused by some external ethnic factor bringing simultaneously alien forms of material culture. Although survey of archaeological evidence has demonstrated rather the gap between the two development phases in the area north of the Middle Danube and isotope analyses produced arguments for an alien intervention, modern archaeology still finds itself on the horns of a dilemma as to whether this evidence can be seen as an expression of discontinuity and whether migratory movements can be read from the archaeological evidence we have. The theory which we intend to contest of the developmental continuity in the northern Danube region and of the mere metamorphosis of the Danubian-Eastern-Germanic into the Thuringian cultural pattern required a closer look at the cultural, geographical and chronological links between the archaeological entities involved. The presence of new outlandish settlers in the earliest phases of inhumation cemeteries that emerge ca. 500, as revealed by foreign archaeological material, does not rule out the possibility that the survival of autochthonous population, however likely, may have been represented there. The change in power and political conditions after the Lombard victory over the Heruli may have resulted namely in a partial depopulation of the country, but as Paul the Deacon indicates, the Lombards were more powerful from that time and their troops took on many warriors from subjugated nations (Paul. Diac. Hist. Lang ), which would suggest assimilation of the remaining aborigines. For the surviving remnants of the pre-lombard population, however, the new acculturation process meant a clear regress, as shown by the break-up of previous settlement structures, decline of high quality pottery production etc. Special position within this problem is occupied by the distinct tumulus Žuráň near Podolí in Moravia. It is located on an elevation about 10 km east of Brno, on an important road leading to the NE and further to the Moravian Gate (fig. 8. 1). The present height of the burial mound is approximately 12 m but according to the nineteenth

393 The Connections between the Region North to the Danube and Northem Europe. Some Aspects of Ethnic and Social Identity of the Fifth Century Elites 393 Fig. 8. Tumulus at Žuráň and examples of finds from chamber grave no. I

394 394 Jaroslav TEJRAL century reports it was once much higher (Poulík 1995). The excavation of Josef Poulík carried out in have shown that the dominant location of this site was used for prominent burials in prehistoric times already, namely in the Eneolithic and the Final Bronze Age. The piled-up mound of later Migration period barrow, which was reinforced by a mighty dry-stone cladding on the perimeter, completely overlaid the older situation. The barrow contained two chamber graves completely plundered in later times (fig. 9). The eccentrically placed and considerably disturbed grave I was unearthed in the nineteenth century already. It contained skeletal remains of at least five horses, fork-shaped iron objects (fig ) and remnants of a wooden beam decorated with triple interlace pattern (fig. 8. 2). Chamber grave II with skeletal remains of two horses, was examined during a research excavation by Poulík in (Poulík 1995; Tejral 2007a; Tejral 2009: ). Whereas the first grave, based on the interlace pattern, falls most probably within the late 6th century and it might be brought evidently into relation with Lombard occupation of the regions north of the Danube, the finds from chamber grave II give a somewhat contradictory evidence. Several artefacts, for example two glass beads (Tejral 2007b: , Taf. 24. b), can be dated to the end of the fourth or early fifth century (fig , 2). To the same period can also be dated the sparse fragments of glass vessels. A part of them come from a thickwalled glass beaker with faceted walls (fig , 2), which can be reconstructed in drawing and identified as a Nyrup-Dybeck-type beaker (type IV after Eldrid Straume), whose main centre of occurrence with 14 specimens (fig. 11) is indisputably situated in Scandinavia (Straume 1987: 34, Taf. 5. 3, 48, 60; Koch 1987: 105, Abb. 21; Gavrituchin 2011: 46, fig. 4A.7 9; Schultze 2002: 29 30, Taf ; Rau 2008: 223, Abb. 2. 7; 6. 7). Seven small fragments of yellowish-green glass with blue Greek letters made in the so-called flash technique (German: Überfangtechnik) most probably belonged to a beaker with round cut medallions of the Straume VIII type (fig ), which is represented by a single specimen already in the Chernyakhov culture and by several other pieces in the rest of Europe, above all in Scandinavia again (Gomolka-Fuchs 1999: 140, Abb. 7. 1; Näsman 1984a: 160; Straume 1987: 39 40, Taf. 9; Abramovicz 1959: pl. 13; Rau 2008: 232, Abb ). Additional examination of glass fragments has revealed that the assemblage included two other glass vessels, probably of Rhine provenance, dating probably from the late fifth or early fifth century. Most of the other finds from chamber grave II, however, are less sensitive in terms of chronology. The talk is of silver bridle bosses (fig ) and four silver rivets with flat discoid head which were used as fittings with cutting weapons or for some other purposes (fig ). They are generally dated to the Merovingian period. Other objects, such as for example the remnants of golden brocade fabric (fig ), golden fitting of a wooden vessel (fig ), elongated oval iron buckles and others, are not very suitable for dating as well. Even though the ivory pyx with old Christian iconography does not find any reliable chronological supports for dating (fig ), its being compared to older late antique works and a more detailed stylistic analysis agree further on with W. F. Volbach, K. Weitzmann and others, who do not class similar pyxes earlier than mainly with the sixth century (Volbach 1962; Volbach 1976: 120, Taf. 97, Nr. 2000; Weitzmann 1972: 35). An interesting evidence of possible date of the latest burial was provided by the anthropological finding. The human skeletal remains which were found in chamber grave II belonged to a female in the age of about 38 years, 158 cm tall, who underwent delivery. She died from an injury to her left pelvic bone (fig ). The indications of beginning healing give proof that she has survived the injury and died shortly thereafter (Drozdová et al. 2009: ). A regular square hole in the pelvis sized 1 x 1 cm corresponds exactly to the shape and dimensions of the point of a Merovingian ango (fig ), which occurs all over the Danube region only in Lombard graves from the mid and the second half of the sixth century onward e. g. from Lužice, Nikitsch, Pottenbrunn or Freundorf (fig ). From the find can be concluded that the female buried in chamber grave no. II was fatally injured at a time when the whole area on the Middle Danube was already occupied and dominated by groups of Lombards (Tobias, Wiltschke- Schrotta, Binder 2010: 293, Abb. 12). Even though the Žuráň tumulus is sometimes being compared to monumental burial mounds such as Ottarshögen near Uppsala or Alt Uppsala (Arrhenius 1987: 441; Arrhenius 1995, ; Arrhenius 2003: ; Ljundquist 2005: ), it is undoubtedly older. However, there are further mounds in Scandinavia and also in western Europe (Arrhenius 1995: ; Müller-Wille 2006: , Abb. 14). The most spectacular there is the mound overlapping presumably the

395 The Connections between the Region North to the Danube and Northem Europe. Some Aspects of Ethnic and Social Identity of the Fifth Century Elites 395 Fig. 9. Plan of the tumulus at Žuráň in the dimensional comparison with the mound of Childerich at Tournai (Müller-Wille 2006) famed sepulchre of Frankish king Childerich at Tournai (Müller-Wille 1997; Müller-Wille 2006: 139; Kazanski, Périn 2005: 287, fig. 1). Some portion of similarity between the mound at Tournai and at Žuráň is given by the presence of horse burials on both sites. However, whilst in the first case three horse burials containing all in all 21 horses were placed outside of the tumulus, at Žuráň six skeletal remains of horses occurred in the chamber grave no. I and two in chamber grave No II which were situated inside of the large barrow (fig. 9). Based on archaic finds such as beads or glass objects it could be supposed that burials began at Žuráň sometime at the turn between the fourth and fifth century already. This all, together with the conspicuous horse burials, provoked an idea that the burial chamber no. II and along with it also the entire burial mound may have been built along the

396 396 Jaroslav TEJRAL Fig. 10. Reconstructions of repaired facetted glass beakers from princely burials from the fifth sixth century. 1 3 chamber grave no. II at Žuráň; 4 Apahida, grave no. 2; 5 Kvassheim; 6 Evebö; 7 Snartemo (Straume 1987; Marghitu 2001) lines of some equestrian nomadic barrows known from the area of Eurasian steppes where horse skeletal remains were also found (Kazanski, Périn 2005; Kazanski, Périn 2007: 32 35). Even though this alternative is not excluded, it is surprising that similar monumental constructions are not known from the Carpathian basin where Hunnic presence and funeral customs are abundantly documented by both sacrificial hoards and inhumation graves (Bóna 1991; Tomka 1986; Tomka 2007: ). The only parallels of a disturbed mighty construction similar to Žuráň are provided by a find from the territory north of the Danube, namely by the tumulus Schmalzberg (fig ) near Laa an der Thaya (Friesinger, Vacha 1987: 77) and maybe also by the excavated but still unpublished barrows at Těšany southeast of Brno. The former find is dated only roughly based on stratigraphy because the central burial chamber was entirely plundered. The latter find lacks any possibility of dating (fig ). Anyway, as shown by the find of wooden beam decorated with interlace pattern from chamber grave I, excavated already in the 19th century, subsequent graves have been recessed into the burial

397 The Connections between the Region North to the Danube and Northem Europe. Some Aspects of Ethnic and Social Identity of the Fifth Century Elites 397 Fig. 11. Distribution of type IV facetted glass beakers (Schultze 2002) mound probably still during the 6th century. The contentual affinity between the finds from this grave and those from burial chamber II, which is reflected in the presence of horse sacrifices, and the aforesaid argumentation based not only on the analysis of ivory pyx but also on the anthropological finding indicate that the female from grave no. II also may have been laid to rest in the grave pit as late as in the sixth century. Even though the old-fashioned artefacts do not rule out that burials at this site were performed in the early fifth century already, their occurrence in the grave, mainly as far as precious glass vessels are concerned, can also be explained in other way. As is indicated by scarce remnants of the original inventory, namely by an ivory pyx or by golden threads from a brocade fabric and, after all,

398 398 Jaroslav TEJRAL Fig. 12. Finds from chamber grave no. II in the mound of Žuráň

399 The Connections between the Region North to the Danube and Northem Europe. Some Aspects of Ethnic and Social Identity of the Fifth Century Elites 399 Fig. 13. The square-shaped defect in the left pelvic bone of the female skeleton from grave no. II at Žuráň, caused by ango; the distribution of Merovingian angos in Lombard graves dated to the mid- and second half of the sixth century

400 400 Jaroslav TEJRAL Fig. 14. Remains of stone construction of the mound at Schmalzberg in Lower Austria (Friesinger-Vacha 1987)

401 The Connections between the Region North to the Danube and Northem Europe. Some Aspects of Ethnic and Social Identity of the Fifth Century Elites also by the extraordinarily demanding construction of the tomb, this place must have been intended for burial of a prominent person, probably somebody of royal descent. This, however, does not rule out that the ancient artefacts may have been part of a royal or queen s treasure (Quast 2011: 135) from which unfortunately only minute fragments remained preserved and which may have been hoarded up over a longer time period. Important is also the fact that analogous depositing of older mended vessels in more recent graves (fig ) is only in Scandinavia documented in at least fifteen cases (Straume 1987: 48 53, 147, table 17), and the same is the case with the facetted glass beaker from royal tomb II of Apahida which was laid aside first in the second half of the fifth century (Marghitu 2001: 154, no ). The exceptional nature of funerary equipment, above all the number of glass vessels, makes to suppose that this tomb contained burial of somebody who maintained special funerary customs other than those which were common with the realm of Lombard burial grounds dated into the sixth century. With regard to the unusual grave goods, the opinion was presented that the person buried here has maintained traditions 401 similar to funerary rites whose analogies occur with grave finds from the sphere of Eastern Germanic populations, but mainly from Scandinavia. Even though historical interpretations are problematic, it is possible that it was grave of a female member of the Herulian nobility, who has maintained her identity in new conditions of the sixth century. It is even not excluded that this female was Silinga, daughter of the defeated Herulian king Rodulf and later third wife of Lombard ruler Wacho, who gave birth to his only son Waltari. This son died under mysterious circumstances after the death of Wacho and after dynastic changes on the Lombard throne. These events may also have been fatal to his mother. Anyway, the occurrence of unusual types of glass vessels and their high number in a single tomb is entirely exceptional in the Merovingian environment of Lombard burial grounds. The facts found, however, do not contradict in any way that the female may have been laid into the grave during the 6th century. The finds can thus help to elucidate the complicated intercultural relations in a period in which the Elbe-Germanic groups led by the Lombards have gained control over the Middle Danube region. Bibliography Abramowicz A., Dąbrowski K., Jażdżewski K., Nosek S. Periode des migrations des peuples // Inventaria Archaeologica. Pologne, Corpus des ansembles archéologiques. II. Łódź, Anke B. Studien zur Reiternomadischen Kultur des 4. bis 5. Jahrhunderts. Beitr. Ur- u. Frühgesch. Mitteleuropas. 8/1 2. Weissbach, Arrhenius B. Skandinavien und Osteuropa in der Völkerwanderungszeit // Germanen, Hunnen und Awaren. Schätze der Völkerwanderungszeit. Ausstellungskat. Nürnberg, Arrhenius B. Regalia in Svealand in early Medieval times // Tor / 1. Arrhenius B. Ottarshögen. 2. Archäologisches. RGA 2. Berlin; New York, Bemmann J. Eine völkerwanderungszeitliche Bestattung aus Epöl, Kom. Esztergom, mit Schwertriemendurchzügen skandinavischer Form // Miscelanea Romano-Barbarica in honorem septagenarii Ion Ioniţă oblata. Honoria. 2. Iaşi, Bemmann J. Mitteldeutschland im 5. Jahrhundert Eine Zwischenstation auf dem Weg der Langobarden in den mittleren Donauraum? // Kulturwandel in Mitteleuropa. Langobarden Awaren Slawen. Bonn, Bemmann J., Hahne G. Waffenführende Grabinventare der jüngeren römischen Kaiserzeit und Völkerwanderungszeit in Skandinavien. Studie zur zeitlichen Ordnung anhand der norwegischen Funde // BRGK Beninger E. Germanengräber von Laa a. d. Thaya (A-Ö) // Eiszeit u. Urgesch Beninger E. Germanenfunde des 5. Jahrhunderts von Wien XXI-Leopoldau // Mannus Bierbrauer V. Neue ostgermanische Grabfunde des 5. und 6. Jahrhunderts in Italien // APA Bliuienė A., Steponaitis V. Wealthy Horsemen in the Remote and Tenebrous Forests of East Lithuania during the Migration Period // The horse and man in European antiquity: worldview, burial rites, and military and everyday life / Archaeologia Baltica. 11. Klaipėda, 2009.

402 402 Jaroslav TEJRAL Bóna I. Der Ausbruch des Mittelalters. Gepiden und Langobarden im Karpatenbecken. Budapest, Bóna I. Das Hunnenreich. Budapest; Stuttgart, Bóna I., Nagy M. Gepidische Gräberfelder am Theissgebiet. I / Monumenta Germanorum Archaeologica Hungariae. 1. Budapest, Čižmář Z. Cholina (okr. Olomouc) // Přehled Výzkumů Čižmář M., Tejral J. Kriegergräber aus dem 5. Jh. n. Chr. in Prostějov-Držovice // Probleme der frühen Merowingerzeit im Mitteldonauraum / Spisy AÚ AV ČR. 19. Brno, Čižmář M, Geislerová K., Rakovský I. Pohřebiště z doby stěhování národů ve Strachotíně // Památky Archeologické Csallány D. Archäologische Denkmäler der Gepiden im Mitteldonaubecken ( u. Z.) // Archaeologia Hungarica. 38. Budapest, Cseh J., Istvánovits E., Lovász E., Mesterházy K., Nagy M., Nepper I. M., Simonyi E. Gepidische Gräberfelder im Theissgebiet. II / Monumenta Germanorum Archaeologica Hungariae. 2. Budapest, Daskalov M. Ein Grab der Völkerwanderungszeit aus Südwestbulgarien (2. Hälfte des 5. Jhs. Anfang des 6. Jhs.) // Archaeologia Bulgarica Drozdová E., Unger J., Smrčka V., Němečková A., Krupka P. Anthropological examination of skeletal remains of a princess buried at the Žuráň barrow (south Moravia, Czech Republic) // Glaube, Kult un Herrschaft. Phänomene des Religiösen im 1. Jahrtausend n. Chr. in Mittel- und Nordeuropa, Akten des 59 Internat. Sachsensymposions. Bonn, Ellågard A. Who were the Eruli? // Scandia /1. Fabech Ch. Neue Perspektiven zu den Funden von Sösdala und Fulltofta. Studien zur Sachsenforschung. 7. Hildesheim, Friesinger H. Bemerkungen zu den frühgeschichtlichen Grab- und Siedlungsfunden von Wien-Leopoldau // Archaeologia Austriaca Friesinger H., Vacha B. Die vielen Vätter Österreichs. Römer Germanen Slawen. Eine Spurensuche. Wien, Gavrituchin I. Cut Glass Beakers within the Context of Studies in the Connections between the South of Eastern Europe and Scandinavia in the Late Period of Roman Influence and the Great Migration Period // Inter Ambo Maria. Contacts between Scandinavia and the Crimea in the Roman Period. Kristiansand; Simferopol, Godłowski K. Die Völkerwanderungszeit in Pommern / Archaeologia Baltica. IV. Polonia. 79. Łódź, Godłowski K. Das Fürstengrab des 5. Jhs. und der Fürstensitz in Jakuszowice in Südpolen // La noblesse romaine et les chefs barbares du III e au VII e siécle. Condé-sur-Noireau, Gomolka-Fuchs G. Gläser der Sîntana de Mureş-Černjachov-Kultur aus Rumänien und der Republik Moldavien // Die Sîntana de Mureş Černjachov-Kultur. Koll. Vor- u. Frühgesch. 2. Bonn, Hårdh B. Uppåkra i volkvandringstiden // Fler fynd i centrum. Materialstudier i och kring. Uppåkra. Acta Archaeologica Lundensia (Ser. in 8 º ). Heather P. Der Untergang des römischen Weltreichs. Stuttgart, Heinrich A. Ein vökerwanderungszeitliches Gräberfeld bei Mitterhof, G. B. Laa an der Thaya, Niederösterreich // Archaeologia Austriaca Kazanski M., Périn P. La tombe de Childéric: un tumulus oriental? // TM. 15. Paris, Kern A. Ein völkerwanderungszeitlicher Töpferofen aus Peigarten, VB Hollabrunn (NÖ) // Gentes, Reges und Rom. Auseinandersetzung Anerkennung Anpassung / Spisy AÚ AV ČR. 16. Brno, Koch U. Die Glas- und Edelsteinfunde aus den Plangrabungen Der Runde Berg bei Urach. VI. Sigmaringen, Lippert A. Völkerwanderungszeitliche Grabfunde aus Schletz und Sigmundsherberg in Niederösterreich // Germania Ljungkvist A. J. Uppsala högars datering // Fornvännen Lotter F. Völkerverschiebungen im Ostalpen-Mitteldonau-Raum zwischen Antike und Mittelalter / Reallexikon Der Germanischen Altertumskunde Erganzungsbande. 39. Berlin; New York, 2003.

403 The Connections between the Region North to the Danube and Northem Europe. Some Aspects of Ethnic and Social Identity of the Fifth Century Elites 403 Lund Hansen U. Small but Important. Glass Fragments from Central Located European Trading Centres (3 rd 6th c. AD) // The Lower Danube in Antiquity (VI c. BC VI c. AD), International Archaeological Con ference, Bulgaria-Tutrakan, Sofia, Marghitu R. Männergrab II von Apahida, Bez. Cluj (Klausenberg, Siebenbürgen / Transylvanien, Rumänien) // Das Gold der Barbarenfürsten. Schätze aus Prunkgräber des 5. Jahrhunderts n. Chr. zwischen Kaukasus und Gallien. Stuttgart, Miletić N. Das frühmittelalterliche Gräberfeld in Rakovčani bei Prijedor // Wissenschaftliche Mitteilungen Bosnisch- Herzegowinischen Landesmuseums. Archeologie. 5. Sarajevo, Müller-Wille M. Prunkgräber der Völkerwanderungs- und Merowingerzeit // Herrschaft Tod Bestattung (Kongreß Kiel). Bonn, Näsman U. Glas och handel i senromersk tid och folkvanderingstid. En studie kring glas från Eketorp-II, Öland, Sverige. Uppsala, 1984a. Näsman U. Zwei Relieffibeln von der Insel Öland // Praehistorische Zeitschrift. 1984b. 59. Périn P., Kazanski M. La tombe de Childéric, le Danube et la Méditerranée // Villes et campagnes en Neustrie / Europe médiévale. 8. Montagnac, Pieta K. Hradiská Bojná II a Bojná III. Významné sídlisko z doby sťahovania národov a opevnenia z 9. storočia // Bojná Hospodárske a politické centrum Nitrianskeho kniežatstva. Nitra, Pohl W. Die Gepiden und die gentes an der mittleren Donau nach dem Zerfall des Attilareiches // Die Völker an der mittleren und unteren Donau im fünften und sechsten Jahrhundert / Berichte des Symposions der Kommission für Frühmittelalterforschung, Oktober 1978, Stift Zwettl, Niederösterreich. Wien, Pohl W. Migration und Ethnogenesen der Langobarden aus Sicht der Schriftquellen // Kulturwandel in Mitteleuropa. Langobarden Awaren Slawen / Instituts Kolloquien zur Vor- und Friihgeschichte. 11. Bonn, Poulík J. Žuráň in der Geschichte Mitteleuropas // Slovenská archeológia Přichystal M., Vachůtová D. Nález hrnčířské pece z doby stěhování národů v Rajhradicích (okr. Brno-venkov) // Sborník příspěvků z I. protohistorické konference Pozdně keltské, germánské a časně slovanské osídlení / Archeologie barbarů Praha, Quast D. Die Langobarden in Mähren und im nördlichen Niiederösterreich ein Diskussionsbeitrag // Archaeology of Identity Archäologie der Identität / Forschungen zur Geschichte des Mittelalters. 17. Wien, 2010a. Quast D. Ortbandzwinge der Goldgriffspathen-Gruppe aus Topornica bei Zamość (woj. Lubelskie) / Monumenta Archaeologica Barbarica. Series Gemina. II. Łódź, 2010b. Quast D. Der Schatz der Königin? Völkerwanderungszeitliche Schatzfunde und weibliche Eliten // Weibliche Eliten in der Frühgeschichte: internationale Tagung vom 13. bis zum 14. Juni im RGZM im Rahmen des Forschungsschwerpunktes Eliten / Römisch-Germanisches Zentralmuseum, Forschungsinstitut für Vorund Frühgeschichte. 10. Mainz, Rau G. Facettschliffgläser in Nord- und Osteuropa // APA Schultze J. Der spätrömische Siedlungsplatz von Wiesbaden-Breckenheim / Kleine Schriften aus dem vorgeschichtlichen Seminar Marburg. 53. Marburg, Staňa Č. Nové nálezy na pohřebišti zdoby stěhování národů v Šaraticích na Moravě // Archeologické rozhledy Stjernquist B. Glass from the Settlement of Gårdlösa, Southern Sweden // Meddelanden från Luds universitents historiska museum Straub P. Angaben zum hunnenzeitlichen ostgermanischen Fundhorizont in Südtransdanubien ausgehend von der Nekropole in Keszthely-Fenékpuszta // Keszthely-Fenékpuszta im Kontext spätantiker Kontinuitätsforschung zwischen Noricum und Moesia. Budapest; Leipzig, Straume E. Gläser mit Facettenschlift aus skandinavischen Gräbern des 4. und 5 Jahrhundert n. Chr. / The Institute for Comparative Resarch in Human Culture. Serie B. Skrifter. LXXIII. Oslo, Szameit E. Zu den Waffen von Wien-Leopoldau // Archaeologia Austriaca Taylor M. Heruler. 2. Historisches. // RGA Tejral J. Mähren im 5. Jahrhundert. Die Stellung des Grabes XXXII aus Smolín im Rahmen der donauländischen Entwicklung zu Beginn der Völkerwanderungszeit / Studie AÚ AV ČR. I. 3. Praha, 1973.

404 404 Jaroslav TEJRAL Tejral J. Morava na sklonku antiky / Monumenta archaeologica. 19. Praha, Tejral J. Archäologischer Beitrag zur Erkenntnis der völkerwanderungszeitlichen Ethnostrukturen nördlich der mittleren Donau // Typen der Ethnogenese unter besonderer Berücksichtigung der Bayern. 2 / Österreichische Akademie der Wissenschaften. Philosophisch-Historische Klasse, Denkschriften Veröffentlichungen der Kommission für Frühmittelalterforschung. 13. Wien, Tejral J. Zur Unterscheidung des vorlangobardischen und elbgermanisch-langobardischen Nachlasses // Die Langobarden. Herrschaft und Identität / Österreichische Akademie der Wissenschaften. Philosophisch- Historische Klasse, Denkschriften 329. Forschungen zur Geschichte des Mittelalters. 9. Wien, Tejral J. Žuráň // RGA. 2007a. 35. Tejral J. Das Hunnenreich und die Identitätsfragen der barbarischen gentes im Mitteldonauraum aus der Sicht der Archäologie // Barbaren im Wandel. Beiträge zur Kultur- und Identitätsumbildung in der Völkerwanderungszeit / Spisy AÚ AV ČR. 26. Brno, 2007b. Tejral J. Langobardische Fürstengräber nördlich der mittleren Donau // Glaube, Kult und Herrschaft. Phänomene des Religiösen im 1. Jahrtausend n. Chr. in Mittel- und Nordeuropa / Akten des 59. Internationalen Sachsensymposions. Bonn, Tejral J. Einheimische und Fremde. Das norddanubische Gebiet zur Zeit der Völkerwanderung / Spisy AÚ AV ČR. 33. Brno, Tejral J. Zum Stand der Langobardenforschung im norddanubischen Raum // Langobardische Gräberfelder in Mähren. I. / Spisy AÚ AV ČR. 39. Brno, 2011a. Tejral J. Cultural or ethnic changes? Continuity and discontinuity on the Middle Danube ca A. D. 500 // The Pontic- Danubian Realm in the Period of the Great Migration. Paris; Beograd, Tihelka K. Knížecí hrob z období stěhování národů u Blučiny, okr. Brno-venkov // Památky Archeologické Tomka P. Der hunnische Fürstenfund von Pannonhalma // AAH Tomka P. Über die Bestattungssitten der Hunnen// Attila und die Hunnen. Historisches Museum der Pfalz, Spe yer. Stuttgart, Turčan V. Nálezy z doby sťahovania národov z Bojnej // Zborník Slovenského národného múzea. 97. = Archeológia Volbach W. F. Zur Lokalisierung frühchristlicher Pyxiden // Variae formae veritas una. Kunsthistorische Studien, Festschrift F. Gerke. Baden-Baden, Volbach W. F. Elfenbeinarbeiten der spätantike und des frühen Mittelalters. Pyxiden mit christlichen Darstellungen. / Kat. Vor.- u. Frühgesch. Altemtümer. 7. Mainz, Weitzmann K. Catalogue of the Byzantine and Early Medieval Antiquities in the Dumbarton Oaks Collection. 3. Ivories and Steaties. Washington, Windl H. Völkerwanderungszeitliche Gräber aus Schletz, BH Mistelbach, Niederösterreich // Archaeologia Austriaca Windl H. Ein bemerkenswerters Fundensemble des 5. Jahrhunderts aus Schletz, VB Mistelbach, Niederösterreich // Fundberichte aus Österreich. 34 (1995). Wien, Windl H. Weitere völkerwanderungszeitliche Gräber aus Schletz, MG Asparn an der Zaya, VB Mistelbach, Niederösterreich // Fundberichte aus Österreich. 35 (1996). Wien, Wolfram H. Grenzen und Räume. Geschichte Österreichs vor seiner Entstehung. Österreichische Geschichte Wien, 1995.

405 The Connections between the Region North to the Danube and Northem Europe. Some Aspects of Ethnic and Social Identity of the Fifth Century Elites 405 Ярослав ТЕЙРАЛ Связи между регионом к северу от Дуная и Северной Европой. Некоторые аспекты этнической и социальной интерпретации памятников знати V в. Резюме Археологические данные о регионе к северу от Среднего Подунавья в V VI вв., открытые главным образом на могильниках, не одинаковы, а наоборот, проявляют достаточное культурное разнообразие. Ранняя группа археологических находок демонстрирует очень близкие черты по всему Среднему Подунавью на протяжении V в. Вторую группу, характеризующуюся преобладанием эльбско-германской лепной керамики и, зачастую, присутствием меровингского оружия, связывают с лангобардами или близкими им этническими объединениями VI в. Характер связей между более древней и более новой группами погребальных памятников имеет определяющее значение для широко обсуждаемого вопроса о континуитете поселений в исследуемом регионе, который оказался значительно более сложным, чем полагали ранее (Bemmann 2008: ; Quast 2010). Появление первой, ранней группы археологических памятников, погребальных обрядов и культур, существенно отличающихся от существовавших ранее, в позднеримское время, относится ко второй трети V в. Это отдельные погребения по обряду трупоположения, группы семейных погребений и более крупные могильники, а также археологические находки на ряде поселений. Попытки решить вопрос об этнической атрибуции этих археологических данных а также о том, кто же был движущей силой исторических событий в Подунавье в V в., сталкиваются с существенными сложностями. Поразительное единообразие археологических находок в трупоположениях в широкой зоне от Среднего Подунавья до регионов, соседних с Балканами, в течение V в. подтверждает существование в этой зоне срав нительно гомогенного культурного круга, развивавшегося параллельно с прото- и раннемеровингской культурой Запада и, несмотря на взаимовли яния и большое число общих черт, сохранявшего свой индивидуальный характер. Поскольку археологические данные о так называемом Дунайском Эльбско-германском культурном круге отражают особенности жизни и моды многочисленных народов, расселявшихся в Среднем и Нижнем Подунавье, зачастую их можно рассматривать как показатель «региональности», а не этничности. Тем не менее, некоторые необычные, но характерные особенности костюма, погребального инвентаря и других вещей, указывающие на осознание своего отличия от современников или на особенности погребальных обрядов, могут, согласно некоторым авторам, указывать на то, что их носители принадлежали к особой общественной, и в этой связи, быть может, этнической группе. Помимо явления, которое в археологическом материале и антропологических находках проявляется как сильное влияние наследия конных варваров, природа этого рода памятников несомненно связана с традициями более ранних восточно-европейских культур. Одной из характерных черт этого внешнего этнического компонента, явно имевшего также социальное измерение, являются отдельные могилы или малые могильники с погребениями, включавшими престижные, социально-значимые детали костюма, дополненные другими симптоматичными элементами погребального инвентаря (рис. 1 4). Особое место занимает небольшая группа находок, демонстрирующих связи с севером Европы, даже если их нельзя считать четко определенной и обособленной группой. Однако свидетельства культурных связей между Средним Подунавьем и северными регионами Европы можно выделить и в более общем хронологически и территориально диапазоне по крайней мере с первой половины V в. (рис. 6 7).

406 406 Jaroslav TEJRAL Как бы ни бросались в глаза находки, определенно демонстрирующие связи соответственно со Скандинавией, польским Поморьем и землями балтов, пока ещё рано связывать их с конкретными этническими элементами. Несмотря на то, что многие вопросы можно решить лишь предположительно, представляется, что этническая ситуация в регионе к северу от Среднего Подунавья оставалась размытой, потому, надо полагать, там находились очень разные этнические объединения. Кроме того, народы и племена, фигурирующие в письменных источниках под конкретными этнонимами, составляли не всё населения данной территории, но, вероятнее его, ведущую общественную прослойку, элиту, господствовавшую над многочисленными группами, включавшими вероятно остатки местных свевов и осколки других племенных объединений, происходивших из разных мест. Согласно письменным источникам, основным фактором власти в Среднем Подунавье были ге рулы, как предполагают, имевшие скандинавское происхождение (Ellagård 1987; Taylor 1999). Где-то здесь, к северу от дунайской границы, герулы под предво дительством прославленного царя Родульфа столкнулись с лангобардами и были обращены в бегство (из последних работ, Pohl 2008: 5 6). Отражением резкого изменения археологической ситуации является вторая, более поздняя группа могильников в регионе к северу от Среднего Подунавья, что представляет собой важную проблему, находящуюся в центре острой дискуссии. На эту перемену указывает возникновение большого количества погребальных памятников с комплексами, которые являются несомненным проявлением эльбскогерманских или «тюрингских» культур, во многом отличающиеся от ранней группы могил (Bemmann 2008: ). Наибольшее количество таких находок распространилось по территории Нижней Австрии над Дунаем и в южной Моравии. В погребальных памятниках этой группы имеется выдающееся количество разнообразной лепной посуды, в основном чашеобразных сосудов, декор которых говорит об их эльбском происхождении. Погребальный инвентарь отличается от предыдущей группы могильников в основном также благодаря отдельным находкам, уцелевшим при ограблении памятников, которое случалось часто. В мужских погребениях обычно находят оружие: мечи обычные меровингские спаты, детали щитов, разные типы пик, копий и метательного оружия, включая несколько выдающихся длинных копий франкского происхождения, известных как ангоны. Часто встречаются разнообразные топоры. Таким образом, несмотря на некоторые общемеровингские вещи, погребальный инвентарь ранней и поздней групп могильников существенно отличается. Данные археологии Моравии и Нижней Австрии противоречат теории о непрерывном развитии местного населения и его истоках. На рубеже V VI вв. имеются сведения о существенных изменениях в размещении населения. Подавляющее большинство погребальных памятников открыты не там, где находились могильники предшествующего времени. Хотя изучение археологических свидетельств и указывает на вероятное отсутствие преемственности между двумя этапами истории региона к северу от Среднего Подунавья, а изотопный анализ дёт основания сделать вывод о вторжении врагов, современная археология до сих пор решет вопрос о том, считать ли эти свидетельства показателем прекращения непрерывного развития, а также насколько имеющиеся археологические данные могут быть отражением миграций. Присутствие новых переселенцев издалека на ранних фазах некрополей с трупоположениями, возникших ок. 500 г., как показывает чужеродный археологический материал, не исключает возможности (пусть маловероятной) присутствия выжившего автохтонного населения. Смена власти и политических условий после победы лангобардов над герулами могла привести к частичному обезлюдиванию страны, но, как писал Павел Дьякон, Лангобарды же стали с того времени мощными, их войско пополнилось многочисленными народностями, которых они победили (Paul. Diac. Hist. lang ), что должно свидетельствовать об ассимиляции уцелевшего населения. Особое место в данном вопросе занимает отдельно стоящий курган Журань близ Подоли в Моравии. Он находится на возвышенности примерно в 10 км от Брно, у важного пути, ведущего на северо-восток и далее к Моравским

407 The Connections between the Region North to the Danube and Northem Europe. Some Aspects of Ethnic and Social Identity of the Fifth Century Elites 407 воротам (рис. 8.1). Под курганом находились две камерные могилы, полностью разграбленные в древности (рис. 9). Необычно расположенная, сильно разграбленная могила I была открыта ещё в XIX в. Здесь найдены костяки по крайней мере пяти лошадей, железное изделие в виде вилки (рис ) и остатки деревянной балки, украшенной тройной «плетёнкой» (рис. 8. 2). В камерной могиле II, исследованной раскопками Й. Пулика в гг., находились скелеты двух лошадей (Poulík 1995; Tejral 2007a; Tejral 2009: ). Если первая могила, судя по орнаменту в виде «плетёнки», вероятнее всего относится к концу VI в. и, очевидно, может быть связана с завоеванием регионов к северу от Дуная лангобардами, находки из камерной моги лы II дают несколько противоречивую картину. Некоторые вещи, например, две стеклянные бусины (рис ), редкие фрагменты кубка с фасетированными стенками (рис ), который можно реконструировать по рисунку как тип Нюруп-Дюбек (тип IV по Э. Штрауме), и фрагменты желтовато-зеленого стакана с синеватыми греческими буквами, в технике наложения цветных слоёв на основу, типа VIII по Штрауме, может относиться к концу IV началу V в. Однако большая часть прочих находок из камерной могилы II менее выразительны с точки зрения хронологии, они датируются меровингским временем в целом. Шкатулка из слоновой кости со старинной христианской иконографией не имеет надёжных хронологических реперов (рис ). Она сопоставима с более древними позднеантичными изделиями, а более подробный стилистический анализ согласуется с выводом В.-Ф. Фольбаха, K. Вайцмана и других исследователей, которые датируют подобные вещи временем не ранее VI в. в целом (Volbach 1962; Volbach 1976: 120, Taf. 97, Nr. 2000; Weitzmann 1972: 35). Интересное данные к возможной хронологии более позднего погребения дают антропологические находки. Остатки человеческого скелета, открытые в камерной могиле II, принадлежали рожавшей женщине возрастом около 38 лет, ростом 158 см. Она умерла от травмы левой тазовой кости (рис ). Признаки начавшегося выздоровления доказывают, что она умерла не сразу после ранения, а через некоторое время (Drozdová et al. 2009: ). Правильное квадратное отверстие в тазе размером 1 x 1 см точно соответствует размеру и форме наконечника меровингского ангона (рис ), которые встречаются в Подунавье повсеместно, исключительно в лангобардских могилах, начиная с середины второй половины VI в. (рис ). Эта находка позволяет заключить, что смертельную рану женщине, похороненной в камерной могиле II, нанесли в то время, когда всё Среднее Подунавье было захвачено и поставлено под контроль группами лангобардов (Tobias, Wiltschke-Schrotta, Binder 2010: 293, Abb. 12). На основании архаичных находок, таких как бусы и стеклянные изделия, можно предположить, что погребения в Журане стали совершать примерно с рубежа IV V вв. В любом случае, находка деревянной балки, украшенной «плетёнкой», в камерной могиле I, раскопанной в XIX в., показывает, что более поздние могилы были впущены в курганную насыпь, вероятно, в течение VI в. Сходство инвентаря, найденного в этой могиле, и находок из погребальной каме ры II, проявившиеся в жертвоприношениях коней, и приведённые выше аргументу, основанные не только на исследовании шкатулки из слоновой кости, но и на антропологических находках, доказывают, что погребение женщины в моги ле II также могло состояться в VI в. Хотя архаичные артефакты не исключают то, что погребения на памятнике совершались уже в начале V в., их появление в могиле особенно дорогих стеклянных сосудов не исключает того, что древние вещи могли быть частью сокровищ «короля» или «королевы» (Quast 2011a: 135), от которых сохранились лишь незначительные фрагменты, и которое накапливались в течение длительного времени. Важно и то, что аналогичная ситуация, когда сосуды со следами ремонта оказались в более поздних могилах (рис ), только в Скандинавии фиксируется по крайней мете в пятнадцати случаях (Straume 1987: 48 53, 147, Table 17). То же самое произошло с фасетированным стеклянным кубком из королевской гробницы II в Апахиде, который был отложен во второй половине V в. (Marghitu 2001: 154, no ). Необычный характер погребального инвентаря, прежде всего большое количество стеклянных сосудов, заставляет предпо-

408 408 Jaroslav TEJRAL ложить, что в гробнице находилась погребение по особому обряду, отличавшемуся от распространённых в лангобардских могильниках VI в. Относительно погребального инвентаря имеется мнение, что захоронение совершено в соответствии с традицией, близкой погребальным ритуалам восточногерманского населения, но главным образом Скандинавии. Хотя любая историческая интерпретация проблематична, возможно, в этой могиле была похоронена представительница герульской знати, сохранявшей своё самосознание в новых условиях VI в. Нельзя исключать и того, что этой женщиной была Силинга, дочь побежденного короля герулов Родульфа, впоследствии третья жена лангобардского правителя Вахона.

409 409 Rostislav TERPILOVSKIY, Gennadiy ZHAROV A Warrior Cemetery from the Turn of the Eras on the River Seym This paper deals with the cemetery from the turn of the eras, located near the modern village of Mutin, in Krolevets district of Sumy oblast (administrative area), and investigated in spring of 2010 by the Left-Bank Expedition of the Institute of Archaeology of the National Academy of Sciences of the Ukraine. Excavations of the site became possible after Aleksandr Lebed discovered by chance a cremation grave in late 2009 (Терпиловский, Билинская 2010). The volume of abstracts of previous Inter Ambo Maria conference contains a preliminary report of this excavation (Terpilovskiy 2010). This compact cemetery is located on the edge of the right side of the Seym river, amidst trees. There, the main river side forms a line of promontories divided by deep ravines. On one of the promontories, there is a fortified settlement separated from the mainland by high steep rampart. Prospect trenches in the settlement area have discovered artefacts from the Scythian period and the Romny culture. The cemetery is located below the settlement, on the upper part of the steep slope, about 20 m high above the flood-plain, mostly within a small platform (fig. 1). Apart from the above-mentioned burial 1 (of 2009), our excavations have discovered 13 cremation graves (including one twin and one destroyed) in the area of 360 square metres. Thus, the cemetery consisted of 15 burials total. Some graves were possibly destroyed about fifty years ago when the trees were planted there. However, the finds were almost absent out of assemblages, so there hardly were more than one or two destroyed burials (Terpilovskiy, Zharov 2012: 301). Most likely, the cemetery of Mutin belongs to a specific cultural group in the middle Seym river, detached under the name of the Khar yevka type of sites by Andrey Oblomskiy and Rostislav Terpilovskiy in early 1990s. The excavations of settlements of Krasnoye 4, Chaplishchi 3, Peresypki 1, and Litvinoichi 3 uncover that they look specifically against the background of the Zarubintsy culture. As the easternmost local variants of this culture, the sites of Khar yevka type still include the most numerous and outstanding western elements. This way, ceramic complex divides into two main groups of different origin. A part of forms of kitchen and table ware are typical of the Zarubintsy culture, especially its middle Dnieper variant. Particularly interesting are the vessels of the second group, which origin is related to the Central European circle, especially the Przeworsk and Jastorf cultures (Обломский, Терпиловский 1994: , рис. 1 9; Обломський, Терпиловський 1994). The cemetery of Mutin is a unique site in several respects both for Central and Eastern Europe. All the graves, urn cremations, obviously belonged to professional warriors of high rank. Especially interesting is twin burial 8, that possibly belonged to brothers-in-arms, with two accumulations of cremated bones. It consisted of a ceramic urn and a bronze bucket standing close to each other and covered with two shield bosses. Burial 8A was accompanied with two spear heads, double-edged and single-edged swords, helmet of a late La Tène type, battle knife, a pair of spurs, and two fibulae. In burial 8Б (within a situla) there were fragments of scabbard for single-edged sword, knife, two spurs, and three fibulae (fig. 2 3). Every burial was accompanied by a representative set of weapons, as a rule with a bent sword and spear head, as well as a shield boss (or a fragment of it). As it has been mentioned above, a small urn of burial 8Б did not contain sword and spear, though in the neighbouring burial 8А there were two swords and two spears. Two swords and two spears were uncovered in burial 7. Burials 4 and 13 contained four spear-heads each, of different types. All the categories of weapons find more or less close analogies in the materials of the Przeworsk and Oksywie cultures, where many weapons have

410 410 Rostislav TERPILOVSKIY, Gennadiy ZHAROV Fig. 1. Mutin. Excavation begins, April 2010 (photo: Anatoliy Karas ) been discovered and later analysed in the generalising monograph by Polish colleagues (Bochnak 2005; Łuczkiewicz 2006). This way, the burials of Mutin contained 21 spear-heads. All the spears were folded in half but two small artefacts. The main bulk of the spear heads have typical relatively long and narrow blade, widest at low, more or less pronounced longitudinal stiffening rib, and short socket. Their overall length is within cm. Most spears belong to Piotr Łuczkiewicz s types F and L (Łuczkiewicz 2006: , , ryc. 36, 37, 41, 42) or Tomasz Bochnak s types 3 and 6 7 (Bochnak 2005: 76 77, tabl. XXII, XXIII, XXV). There also are single finds of small spear-heads cm long, including artefacts with a loop at the low blade (buri al 10), asymmetrical blade (burial 3), and doublebarbed spears with long socket (burial 4 and probably 13). All the swords are deformed, basically folded several times together with scabbard to form a flat zigzag or spiral. The only single-edged sword from burial 8A is 82.5 cm long, with the blade 5.2 cm wide at the top, and a handle with side holes to rivet appliqués. Extant iron parts suggest that its scabbard consisted of two wooden panels fastened with cross-wise C-shaped clamps, partly with fine rings for hanging, and flattened end-piece. The form of the handle makes this artefact comparable with single-edged swords in the Jastorf and Przeworsk cultures, of type W. IIIВ (Łuczkiewicz 2006: 68) or type F (Bochnak 2005: 65, tabl. XIX). All other swords (14 specimens) are long double-edged, of La Tène type, with aslant shoulders, slightly narrowed and rounded end, and a little bit different size, proportion, and cross-section of the blade. Most of them belong to Łuczkiewicz s type III4 and possibly III5 (Łuczkiewicz 2006: 39 41, ryc. 8 9) or Bochnak s types 2 5 (Bochnak 2005: 25, tabl. II V). Their overall length is cm, with the handle cm long and the blade cm wide at the top. In most cases, there are extant ball-shaped pommels, bell-shaped (bulbous) guards, and fragments of scabbards in the

411 A Warrior Cemetery from the Turn of the Eras on the River Seym 411 Fig. 2. Mutin. Burial 8A. Grave goods: 1 7 iron, 8 black-burnished ceramic vessel (drawing: Terpilovskiy)

412 412 Rostislav TERPILOVSKIY, Gennadiy ZHAROV Fig. 3. Mutin. Burials 8A (1 6) and 8Б (7 20). Grave goods: 1 14, iron, 15 bronze situla with iron handle and loops (drawing: Terpilovskiy)

413 A Warrior Cemetery from the Turn of the Eras on the River Seym form of a thin iron badge, loops, openwork clamps, and end-pieces. Extremely interesting are the finds of several scabbards decorated in opus interrasile style. This way, there is honeycomb decoration on top of the scabbard from burial 3 and network design on the top of the scabbard from burial 8A. The Mutin scabbards generally belong to Łuczkiewicz s types 5 and 6 (Łuczkiewicz 2006: 62 65, ryc. 17) or Bochnak s types VII and VIII (Bochnak 2005: 40, tabl. XIV). Shield bosses or their fragments have been discovered in every burial, usually in the top of the urn, with the dome up. Some artefacts are heavily deformed and poorly preserved. Burials contained several badges / rivets, supplemented with a fragment of flange in burial 9, obviously following pars pro toto principle. Their domes are sub-cylindro-conical, sometimes with pronounced apex; those of the finds from burials 1 and 3 terminate with a longer spike with flattened tip. As a rule, such a boss was riveted to the shield with six, rarely eight large (up to 6 cm in diameter) rivets. The number of smaller badges (less than 2 cm in diameter) reached 12. Burial 3 contained spiked shield boss with 12 rivets and shield handle with figural ends. The main bulk of shield bosses from Mutin probably belonged to types 7, 10, and 11 in Bochnak s classification of the Przeworsk shield bosses (Bochnak 2005: , ryc. 35, tabl. XXXIV XXXVI). In most cases, the complex of weapons was supplemented by spurs and battle knives. Spurs of the same type (18 specimens) with long spike and large convex caps on the ends of yoke belong to Jerzy Ginalski s sub-group А (Ginalski 1991: 55 56). In most cases they formed pairs. Battle knives are up to 17 cm long, as a rule with narrow grooves along the back on either side of the blade, a rivet, and a profiled socket for a wooden shaft. Their scabbards had figural end-pieces of thin iron, discovered in graves 4, 5, 8А, and 12. Three larger endpieces of scabbards of knives or spears originate from burial 13. Other iron ware includes numerous spear-butts, parts of waist or shoulder belts (rings and clasps), and late La Tène fibulae. The exceptions are bronze ring and clasp, as well as two iron razors with curved blades (burials 2 and 6) and mid La Tène fibula from burial 7. Iron helmets from burials 4, 5, 8A, 12, and 13 are unique for both Eastern and Central Europe. In burials 8A and 12, they played a role of urns placed into earthenware vessels. These helmets generally belong to the late Eastern Celtic type. Every helmet 413 consists of three elements: hemispherical bowl, front (with or without peak), and neck-guard, as well as cheek-pieces hinged to the front side. There were loops for a chinstrap on the inner side of cheek-pieces and neck-guards. Cheek-pieces and neck-guards from burials 12 and 13 are decorated with convex ornament. Prestige of the burials is underlined by various vessels used as urns. Burials 9, 12, and 14 had created remains placed in black-burnished ceramic vessels similar to the finds in Khar yevka type of sites. In burials 1 and 4, there were extra metal vessels put in ceramic vases. These metal vessels were urns in 11 burials: eight situlae of various forms (burials 1, 2, 4, 5, 8Б, 10, 11, 13), two low basins (burials 3 and 6), and a bi-metallic cauldron (burial 7) (fig. 4 6). The lack of direct parallels to the mentioned vessels in Eastern Europe forces the one to search for analogies in Eggers classical study of the late La Tène and Roman imports in Germania Libera (Eggers 1951). Actually, it contains positive parallels to most of our finds. It might be supposed that the set of metal vessels in Mutin divides into two groups by origin. Celtic artisans probably made the most of the situlae with iron loops and handles, and a cauldron with iron top and two ring handles. Roman workshops probably produced two elegant situlae with bronze attachments of handles and handles, as well as basins with convex bottom. The first group probably includes four situlae of various sizes from burials 1, 4, 5, and 13, of type Е.22 (Eggers 1951: 161, taf ), all of rough workmanship, carinated body, iron handle and loops (fig. 4. 4; 6. 1, 4). A special place among them is occupied by deformed situla from burial 4, with vertical holes for rivets to fasten loops in contrast to their usual horizontal location. Two more buckets of barbarian workmanship from burials 8Б and 10 do not meet with exact analogies in the mentioned catalogue. Some of these vessels were intensively used during a long time, as it was suggested by traces of repair on finds from burials 1, 4, 5, 8Б, and 10. The same group probably includes bi-metallic cauldron of type Е.5 (fig. 5. 1). According to Eggers, such artefacts are simplified replicas of luxury Celtic cauldrons of type Gundestrup (Eggers 1951: 40, 166, Taf. 2. 5). The second group of vessels is presented particularly by a thoroughly made bucket from burial 2 (fig. 4. 1), close to situlae of type Е.20, with typical trapezoid bronze loops (Eggers 1951: 40, 161,

414 414 Rostislav TERPILOVSKIY, Gennadiy ZHAROV Fig. 4. Mutin. 1 burial 2; 2 burial 3; 3 burial 4; 4 burials 5 and 6 (photo: Zharov) Taf ). However, the shape of handle attachments of the Mutin situla is a little different. For example, there is a vessel with similar loops from a treasure in Podmokly in modern Czech Republic (Droberjar 2006: 51, obr ). Large situla from grave 11 has the body of noble form, with gently curved neck and carefully treated details (fig. 6. 2), though it does not find exact parallels in the mentioned classification. According to Eggers catalogue, bronze basins of type Е.67 from graves 3 and 6 (fig. 4. 2, 4), probably made in northern Italy, similarly to other Roman imports, were widespread in Germania Libera in the late La Tène and early Roman period. Eggers has listed 17 finds, concentrated in the lower Elbe area, where such basins were often used as grave urns in cremation graves of the Jastorf culture (Eggers 1951: 41, Taf ). Imported bronze situlae, basins, and cauldrons, similar to the finds from Mutin, are known in modern Poland. According to Teresa Dąbrowska, a cauldron of type Е.5 is discovered in the Przeworsk

415 A Warrior Cemetery from the Turn of the Eras on the River Seym 415 Fig. 5. Mutin. 1 burial 7; 2 burial 9; 3 burial 8; 4 a fragment of burial 8А (photo: Zharov) burial А2 from the late pre-roman period and buckets Е and basins Е.67 in graves from phase А3, suggesting two waves of Roman vessel penetrated into the Barbaricum (Dąbrowska 1988: ). The funeral rite in Mutin meets with most close parallels amidst such cultures of the north-western circle under La Tène influence as the Przeworsk and Oksywie cultures. They are typical of cremations in urns, accompanied with ritually folded or broken weapons (especially spear-heads and swords). According to Łuczkiewicz s calculations, isolated finds and sets of various weapons are documented in about two hundred Przeworsk and one hundred Oksywie graves from the pre-roman period, with imported bronze vessels used as urns for cremations in ten weapon graves (Łuczkiewicz 2006). This way, a set of funeral equipment including sword, spear, and shield is quite typical of the Przeworsk weapon graves, reaching 12 to 19 % in different periods (Bochnak 2005: ). Geographically most close analogies to the sword scabbards in opus interrasile style and other weapons are located as far as the upper Dniester area. In this connection, one should remember burial 25 in the cemetery of Siemiechow, with the only in Poland La Tène helmet used as an urn (Jaźdźewska 1994: ). Since every burial in the cemetery of Mutino was accompanied with numerous goods, these assemblages possess reliable dates. For example, the set of weapons and warrior s equipment is generally typical of the end of the pre-roman period. Swords (with scabbards), spear-heads, and shield-bosses find numerous analogies in the Przeworsk and Oksywie burials from phases А2

416 416 Rostislav TERPILOVSKIY, Gennadiy ZHAROV Fig. 6. Mutin. 1 burial 10; 2 burial 11; 3 burial 12; 4 burial 13 (photo: Zharov) А3. Moreover, the combination of the listed above types of swords, their scabbards, spear-heads, and shield bosses is typical of phase А3 (Łuczkiewicz 2006: ). Ginalski s sub-group A of spurs also dates back to the late pre-roman period, mostly to phase A3 (Ginalski 1991: 55 56). It is interesting to note that pairs of spurs are comparatively rare finds in Poland: they appeared in seven graves of the Przeworsk and four of the Oksywie culture respectively (Łuczkiewicz 2006: 146). On the contrary, they are discovered in eight burials in the cemetery of Mutin. Winged clasps of Józef Kostrzewsky s type К.49 (Kostrzewski 1919: Abb. 49) and solid rings were used in waist and shoulder-belts for swords from the turn of phases А1 / А2 to the late pre-roman period (Dąbrowska 1988: 30; Bochnak 2005: 57 58). The late La Tène fibulae of Kostrzewski s type N form the most distinctive chronological indicator (Kostrzewski 1919: abb. 24). The burials of

417 A Warrior Cemetery from the Turn of the Eras on the River Seym 417 Mutin contained 13 artefacts of the kind. Such artefacts, as well as similar brooches of type М (with high cord), existed in the late pre-roman Iron Age assemblages under the La Tène influence, mostly of the Jastorf, Przeworsk, and Oksywie cultures, less in the Zarubintsy cultures. There are similar fibulae from the cemetery of Dolinyany, which is usually related to the Poieneşti-Lukaşevka culture (Смирнова, Мегей 2000: рис ). In Dąbrowska s opinion, fibulae of type N are typical of the Przeworsk culture during the second horizon of phase A3, and the latest are big artefacts with openwork foot-plate (like some fibulae from Mutin). This horizon dates from the last but final decade of the first century BC to the early-second decade of the first century AD (Dąbrowska 1988: 61 62). Typologically earlier is the fragmented fibula of mid La Tène construction from grave 7: it is characteristic of phase A2. The same assemblage contained two swords with fragments of scabbards of Bochnac s type III and bi-metallic cauldron of type Е.5, which are characteristic of phase А2 of the Przeworsk culture (Dąbrowska 1988: 29, tabl. III). The great bulk of metal vessels in Mutin is related to the second wave of Roman imports, as numerous finds of fibulae of type N suggest. Perhaps burials in Mutin cemetery are generally synchronous within the pre-roman period А3: they were made in the period from the mid-first century BC to the early first century AD, most likely on the turn of the eras. It is worth paying attention that it was the time when the grave goods were deposited and these artefacts became a part of the dead culture. However, several items were intensively used for years before that, as it comes, for example, from the traces of repair on buckets from burials 1, 4, 5, 8Б, and 10, or heavily worn loops of the situla from burial 2. Therefore, the bronze ware from Mutin was quite possibly made a little earlier than the narrow-dated fibulae, about the mid-first century BC. Burial 7 probably dates back to the same period, thus forming a kind of relic from the previous period in Mutin. In contrast to metal artefacts, we do not know close analogies to ceramic urns from the cemetery of Mutin among the cultures under the La Tène influence. Nevertheless, the similarity of polished vessels from Mutin and tableware of western tradition in the above-mentioned settlements of Khar yevka type in the basin of the river Seym is beyond any doubt: they have common form, proportions, faceted rim, decoration (with vertical combing), bosses, and X-shaped handles. We have already mentioned that bronze and ceremonial ceramic vessels used as urns and the complex of military equipment with swords, helmets, and in most cases spurs supply evidences of high status of the buried persons. The relatively small number of burials with prestigious goods and isolation of the cemetery obviously indicate that warriors (of prince s retinue) were buried there. The western origin of the warriors who came to the Seym area is indicated primarily by the complex of weapons typical of cultures of the late La Tène circle. It is known that such arms were used by ethnic Celts and other tribes under the La Tène influence, adding their own tribal / national bladed weapons, like Geto-Dacian sickle-shaped sica sword or Germanic straight single-edged broadsword. Single-edged sword from burial 8, typical of the Oksywie and to lesser extent Przeworsk culture, allows us to suppose Germanic origin of this retinue, or at least some of its members (Terpilovskiy, Zharov 2012: ). Obviously, these warriors possessed metal vessels, later used as urns. Such warriors set of vessels for collective feasts included a cauldron, two basins, a big ceremonial situla, as well as several buckets. Ceremonial ceramic vessels from burials 1, 4, 8А, 9, 12, and 14, also typical of the sites of type Khar yevka, could indicate the inclusion of the migrants into the structure of local tribes. It is quite probable, because the people from the Khar yevka type of sites established contacts with the Central European populations from the very beginning. The synchronism of the graves is possibly an evidence that these warriors were killed in a single battle. At the same time, they probably triumphed: their tribesmen buried them following all the details of the ritual, not carrying about money, in a special cemetery. Although cremations in urns of imported metal vessels, accompanied with weapons, are known in the Przeworsk and Oksywie cultures, similar cemeteries of warriors are not known there. A kind of analogy for Mutin cemetery could be weapon graves from horizon (or culture) Grossromstedt in Middle Germany and Czech Republic (Droberjar 2006). Nevertheless, the cemetery of Mutin is unique because it consists of only warriors graves from the final stage of the pre-roman period and, moreover, it was located on the eastern extremity of the area occupied by cultures under the La Tène influence.

418 418 Rostislav TERPILOVSKIY, Gennadiy ZHAROV Bibliography Bochnak T. Uzbrojenie ludności kultury przeworskiej w młodszym okresie przedrzymskim. Rzeszów, Dąbrowska T. Wczesne fazy kultury przeworskiej. Chronologia zasięg powiązania. Warszawa, Droberjar E. Planaňská skupina grossromstedtské kultury. K chronologii germánských nálezů a lokalit v Čechách na sklonku doby laténské a počátkach doby římské. // Archeologie Barbarů Sbornik přispevků z I. protohistorické konference Pozdně keltické, germánské a časně slovanské osídleni. Praha, Eggers H. J. Das römische Import im freien Germanien. Hamburg, Ginalski J. Ostrogi kabłąkowe kultury przeworskiej. Klasyfikacja typologiczna // Przegląd Archeologiczny Jaźdźewska M. Ciekawsze zespoły grobowe ze śladami wpływów celtyckich na cmentarzysku kultury przeworskiej w Siemiechowie, woj. Sieradzkie // Kultura przeworska. 1. Lublin, Kostrzewski J. Die ostgermanische Kultur der Spätlatènezeit / Mannus-Bibliothek. 18. Leipzig; Würzburg, Łuczkiewicz P. Uzbrojenie ludności ziem Polski w młodszym okresie przedrzymskim / Archaeologia Militaria. II. Lublin, Terpilovskiy R. A Cremation from the Turn of Eras and the Sites of Khar yevka Type in the Middle Seym Area // Inter Ambo Maria. Contacts Between Scandinavia and Crimea in the Roman Period (Abstracts). Simferopol, Terpilovskiy R., Zharov G. A Warrior Cemetery from the Turn of the Eras on the River Seym // Inter Ambo Maria. Northern Barbarians from Scandinavia towards the Black Sea. Simferopol, Обломський А. М., Терпиловський Р. В. Посейм я у латенський час // Археологія Обломский А. М., Терпиловский Р. В. О связях населения Центральной Европы и востока Днепровского Левобережья в латенское и раннеримское время // Kultura przeworska. 1. Lublin, Смирнова Г. И., Мегей В. Ф. Могильник типа Поянешть-Лукашевка у с. Долиняны (раскопки 1985, 1987, 1988 гг.) // Stratum plus Терпиловский Р. В., Билинская Л. И. Трупосожжение знатного воина рубежа эр на Сейме // Германия Сарматия. II. Калининград; Курск, Ростислав ТЕРПИЛОВСКИЙ, Геннадий ЖАРОВ Дружинный могильник рубежа эр на Сейме Резюме Весной 2010 г. Левобережной экспедицией Института археологии НАН Украины был исследован могильник Мутин на р. Сейм (северовосточная Украина). Раскопки памятника стали возможны после случайной находки кремационного погребения А. Лебедем. Кроме него, в ходе раскопок на площади около 360 м 2 обнаружено ещё 13 захоронений с трупосожжениями (в том числе одно парное и одно разрушенное). Как выяснилось, Мутинский могильник в ряде отношений уникален. Все захоронения, урновые кремации, очевидно, принадлежали воинампрофессионалам высокого ранга. В большинстве случаев урнами служили металлические сосуды: восемь ситул различных размеров и форм (в том числе типов «Эггерс 20, 22»), два низких таза (типа «Эггерс 67») и биметаллический котёл типа «Эггерс 5». Каждое погребение сопровождалось представительным набором оружия, включавшим, как правило, умбон, согнутые обоюдоострый меч и наконечник копья. В отдельных случаях в погребении находились по два меча и копья. Обычно комплекс вооружения дополнялся шпорами, бое-

419 A Warrior Cemetery from the Turn of the Eras on the River Seym выми ножами, втоками копий, частями портупей или поясов (кольцами и застежками), фибулами и др. Уникальными не только для Восточной, но и для Центральной Европы, являются находки в пяти погребениях железных шлемов. Погребальный обряд Мутинского могильника находит ближайшие соответствия среди таких латенизированных культур северо-западного круга, как пшеворская и оксывская. Набор оружия и воинского снаряжения из погребений могильника в целом характерен для этих же культур финала предримского времени. Наиболее узко датируются фибулы позднелатенской схемы типа N (13 экземпляров), распространенные в комплексах второго горизонта фазы А3 пшеворской культуры по Т. Домбровской. Этот горизонт датируется предпоследним десятилетием I в. до н. э. началом второго десятилетия I в. н. э. Типологически более ранней является единственная фибула среднелатенской схемы из погребения 7, типичная для фазы А2. Вместе с тем, керамические урны Мутинского могильника находят аналогии среди столовой керамики «западной» традиции памятников типа «Харьевка» бассейна реки Сейм. 419 Бронзовые и парадные керамические сосуды, использованные в качестве урн, а также комплекс воинского снаряжения свидетельствуют о высоком социальном статусе погребенных. Сравнительно небольшое число погребений с престижным инвентарем и изолированный характер могильника, очевидно, указывают, что здесь захоронены представители военной дружины. На «западное» происхождение дружинников, прибывших на Сейм, в первую очередь, указывает комплекс вооружения, типичный для позднего латена. Однолезвийный меч из погребения 8 позволяет предположить германское происхождение дружины (или, по крайней мере, отдельных ее членов). Парадные керамические сосуды типа Харьевка, использованные в погребениях могильника, могут свидетельствовать о включении мигрантов в структуру местных племён. Хотя погребения с оружием достаточно распространены в пшеворской и оксывской культурах, подобные дружинные могильники там не известны. Определенной аналогией Мутинскому могильнику могут служить погребения воинов горизонта (или культуры) Гроссромштедт рубежа эр в Средней Германии и Чехии.

420 420 Aleksandr VASIL YEV Thin-Walled Glass Beakers with Wide Mouth and Decoration of Cut Ovals (Eggers 223, 224, 226, 228) Cut decoration consisting of various combinations of circular facets and carved lines was typical of several groups of glass vessels from the Late Roman period, distributed between the Baltic and the Black Sea. First and foremost, there were hemispherical bowls Е 216 and cylindrical beakers Е 230. An intermediate position between the two is occupied by a heterogeneous group of glass vessels belonging to Hans Jürgen Eggers types 220, 221, Most of them have rim diameter bigger, though inconsiderably, than mouth dia meter. It is a formal reason to term this type of vessels bowls, as Eldrid Straume did. However, in the most part of publications these vessels are called beakers to underline their simil arity with vessels of type Е 230 rather than type Е 216. Since their proportion separates them from other thin-walled beakers with analogous decoration, I suggest the name of beakers with wide mouth or wide beakers for short. The discovery of a vessel of the type in grave 139 in the cemetery of Neyzats supplies us with every reason to include this ware into the circle of various antiquities from the Roman period, distributed throughout the territory between the Baltic and the Black Sea. The publisher of the Neyzats vessel has already supplied a group of analogies for the beaker under analysis in Scandinavia and the Chernyakhov culture area (Храпунов 2004: 303). In this paper, I will try to enlarge the circle of possible analogies and to clarify their chronolo gical position. The main stages of research of glass beakers with wide mouth and decoration of cut ovals Еggers has selected a heterogeneous group of cut glass bowls Е The most of such finds in his corpus are a single specimens, from phase С2. Assemblages from Sigerstad (bowls Е 221, 223), Skjorringen (Е224), and Ganzkow (Е 220) could also be from the next phase D1 (Eggers 1951: 180). The form and decoration of types are most close within this group. Eggers paper discussing the chronology absolute in the Roman period in Germania Libera is a little bit confusing with the question of interest. In the main body of the text discussing the goods typical of phase С2, vessels Е and Е are considered different chronological indicators, no. 61. b c and 62 respectively. The illustration to this list combines, under no. 61, vessels Е 216 (as а), Е 229 (as b), and Е230 (as с). The table of mutual occurrence includes beaker Е 229 from Varpelev into the general chronological indicator no. 61 with bowls Е 216, though the plate misses assemblages with vessels Е 230, the same as the cluster related to phase С3 (Eggers 1955: 202, Abb a c; Abb 6). This way, Eggers has attributed types 229 and 230 of beakers to phase С2 without reliable reasons, most likely by intuition. An important place in Günter Rau s observations is occupied by an assemblage from Sigerstad with two different glass beakers with cut decoration (Е 221 and Е 223), which allows the one to determine the synchronism of the continental finds with analogous vessels from Redlinо (Е 226) and Ganzkowо (Е 220) (Rau 1972: 129). The same group includes low thick-walled vessels, decorated with deep cut ovals, from Sakrau II and Varpelev (Е 229). Cut beakers of variants Sakrau II, Ganzkowо, and Kowalk form a single horizon of finds to replace the horizon of glass bowls, as a rule accompanied with fine bronze buckets of type Hemmoor. Moreover, as Rau has underlined several times, these metal vessels never appeared in the same assemblages with cut beakers (Rau 1972: 161, 164). The horizon of cut beakers includes an earlier position (ca. AD ), occupied by wide beak-

421 Thin-Walled Glass Beakers with Wide Mouth and Decoration of Cut Ovals (Eggers 223, 224, 226, 228) ers of variants Sakrau II and Ganzkowо. Elongated beakers of variant Kowalk generally date back to ca. AD When the Great Migration period began, they were replaced by conical beakers (Rau 1972: 164, ). In Rau studied late antique beakers in Northern and Eastern Europe an museums and developed a new interpretation of these materials with a special view on the typological classification. He has divided cut beakers into types Varpelev, Kowalk, and Dybeck / Lund. The type Varpelev, of our present interest, includes vessels of very different forms, decoration, colour, and thickness of glass. Rau has dated type Varpelev to the first half, type Kowalk to the middle / second half, and type Dybeck/Lund to the second half of the fourth and early fifth century (Rau 2008: ). Ulla Lund Hansen has also united types Е 221 and into a single group in her corpus of the Roman imports in Scandinavia. She has dated the Sigerstad grave (Е 221, 223) to phase C1b С2, the finds from Store Heddinge / Skjorringen (Е 224) and burial 5 in Vrangstrup (Е ) to phase С2 (Lund Hansen 1987: ), and phase С2 in general to ca. AD 250/60 310/20 (Lund Hansen 1987: 30, 38, Fig. 10, 12). As for the Rau s chronology, Lund Hansen has pointed out that his dates are too young in comparison with those suggested by Kazimierz Godłowski, which she thinks are correct (Lund Hansen 1987: 78). In the study of Scandinavian graves from the fourth and fifth century containing cut glass beakers, Eldrid Straume has analysed the vessels of our present interest as a specific group of bowls (Е 221, , 227, 228), different from beakers (Е ), which in their turn divide into ten types with series and variants. The researcher discusses all the bowls as two sub-groups with different ornamental patterns (Straume 1987: 42 43). Sub-group I incorporates four vessels decorated with lines of polished ovals, of different form, size, and thickness of walls: bowls from Sigerstad (Е 223), Kobbeå (Е 228), Lærkenfeld (227), and grave a in Varpelev (Е 229). As parallels to the vessels of sub-group 1 in the Continent, Straume cites vessels from Redlinо (Е 226) and Zakrzów / Sak rau II (Е 229). The researcher has established the chronology of sub-group 1 within phase С2 (its later period?) D1, but the reasons seem vague. The Sigersted assemblage has been dated to С3 D1 (С3), though it did not contain pronounced chronological goods but two beakers. The 421 Kobbeå grave has been dated by Straume to the Late Roman period (С2 С3) because of handmade vessels (Srtraume 1987: 119), though this assemblage also did not contain other finds capable of clarifying its chronology. The Lærkenfeld assemblage has been dated to D (D 1) (Straume 1987: 119). Apart from the beaker, there was an iron shield boss of Jørgen Ilkjær s type 7, most typical of phase С2/С3 (Ilkjær 1990: 330, Abb. 199; Godłowski 1994: 176, Abb. 5). Straume has dated the grave in Varpelev to phase С3/D1 (Straume 1987: 123). Not coming deep into the complex problem of the chronology of this assemblage, I think it necessary to mention two important points. First, she actually ignores the coin found in this burial (aureus of Probus from AD ) and an analogy to the beaker from the princely complex from phase С2 in Zakrzów / Sakrau II. Second, she uses incorrect analogies to the belt set from Varpelev, namely the finds from Untersiebenbrunn and Jakuszowice for the small buckle and zoomorphic design of footplates of Northern European cross-shaped fibulae (kreuzformiger Fibeln) from the late fourth and early fifth century for the decoration of the big buckle tongue. Sub-group II includes vessels with more homogenous morphology and more complicated decoration: from Sigerstad (Е 221), ornamented with combination of diamonds, horizontal bands, strokes and cut narrow ovals; from Skǿrringen (Е 224), decorated with alternating horizontal lines and lines of wide and narrow ovals; and from burial 5 in Vrangstrup (Е ), decorated with lines of wide ovals within pentagonal cells of engraved lines. The continental analogies to vessels of this group include much more finds, from Ganzkowо, Frunzovka, burial 9 in Kosanovo, burial 82 in Gavrilovka, and Stolac. As for the chronology of sub-group II, apart from the above-discussed assemblage in Sigersted, Straume has dated Skjorringen to the late С2 and burial 5 in Vrangstrup to the late С2 С3 (late С2?) (Straume 1987: 121, 124). The closest analogy to the goods in grave in Skjorringen (fibulae Makeprang III and IX, the composition of the necklace) is among the goods from recently published rich grave in Maglebjerg, from the early stage С2 (Borby Hansen 2010: 193). In my point of view, the chronology of burial 5 in Vrangstrup could be clarified by a gold foil with impression of a coin of the emperor Probus (AD ), and a gold

422 422 Aleksandr VASIL YEV finger-ring. A similar finger-ring was discovered in Norway within an expressive assemblage from stage С2 containing a fibula with high catch-plate Makeprang II, glass bowl Е 204, and bronze basin Е 106 (Straume 1998: , Abb 7). Therefore, we can be sure that burial 5 in Vrangstrup dates back to the late stage С2. Therefore, Straume s chronology of assemblages with glass cups seems too long and too young. Moreover, the typology proposed by the researcher does not seem good, because its classification units combine very different vessels. Classification of glass beakers with wide mouth and cut decoration I suggest the following variants of these vessels. 1. Varpelev-Zakrzów variant (Е 225, 227, 229). Wide and low beakers with medium-thick or thick walls and two or rarely three lines of wide oval and circular cuts. There is one case with ornamentation of cut horizontal lines. As a rule, top line consists of oval and bottom line of circular cuts. 2. Sigerstad-Neyzats variant (Е 223, 224, 226, 228). Wide and slender beakers with thin flaring walls, circular bottom, out-turned rim, and decoration of a few (two or three) lines of oval cuts. The cuts in the lower line are sometimes circular; there are vessels with carved horizontal lines between the lines of fa cets. 3. Sigerstad-Ganzkowo variant (Е 220, 221). Analogous bowls, ornamen ted with a belt of diamonds in the middle and a line of narrow ovals cut below. Sometimes there are alternating diamonds with inscribed ovals and the diamonds filled with oblique net of incised lines. 4. Vrangstrup-Iacobeni variant. Analogous bowls, with the ornamentation based on two rows of cut pentagons with inscribed ovals, forming a honeycomb pattern. Additional elements include belts of narrow ovals placed horizontally or vertically. The subject of my present study are the beakers of variant Sigerstad-Neyzats, including three distinctive series with specific ornamental patterns. 1. Sigerstad series (Е 223, 226, 228). The only decoration of these vessels are wide cut ovals (height: 2 to 2.5 cm, width: 1.2 to 1.8 cm) or circles (diameter ca. 2 cm) forming three or two lines (see the catalogue, nos. 1 7). 2. Skjorringen series (Е 224). These vessels have a line of vertical narrow ovals (height: cm, width: cm) at the middle of the body, limited by cut horizontal double line above and below. Higher and lower than these lines, there are rows of wide or narrow ovals (catalogue nos. 8 9). 3. Häven series. These vessels are decorated with three lines of cut narrow ovals, chequered, close to each other, and a line of circular cuts near the bottom (catalogue nos ). Both Eastern European finds of the beakers under analysis (Velikaya Bugayovka and Neyzats) belong to series Sigersted (fig. 5. 1). Its typical decorative pattern was continued by the ornamentation of beakers Е 230 (fig. 5. 2). Although wide beakers of variant Sigersted-Neyzats are ornamented with two or three rows of separated oval cuts, all the beakers of type Е230 I know (the only exception will be discussed below) are decorated with four lines of ovals. The same ornamentation appears on thick-walled conical beakers from the Great Migration period, united by Igor Gavritukhin into variant Hogom b, within series Goroshevtsy of type Hogom (Гавритухин 1999: 49). By the way, the body of the Hogom beaker is decorated with five lines of ovals (fig. 5. 3). So far wide beakers of series Skjorringen and Haven (fig. 5. 4, 8) are discovered only in the Baltic Sea area. However the total number of finds is too small to make the final conclusion concerning their distribution. In this regard, important is the beaker from grave 78 in the Crimean cemetery of Druzhnoye (Храпунов 2002: 181, рис ), of the form corresponding to type Е 230, but ornamented with lines of narrow and wide cut ovals alternating with double horizontal lines (fig. 5. 5), i. e. its decorative pattern corresponds to series Skjorringen. Such a decoration is not documented among Scandinavian finds of beakers Е230. In the Migration period, such decoration, typical of wide beakers of series Skjorringen, appeared among thick-walled beakers (fig. 5. 6) of Gavritukhin s series Bremsnes of type Lugi (Гавритухин 1999: 53). Decorative pattern of the vessels in series Haven is not restricted to wide beakers. It initially appeared on globular bowls Е 216 (fig. 5. 7), and later on thick-walled conical beaker from Mæle (fig. 5. 9) (Straume 1987: Taf ). Decorative pattern of beakers from Haven and Neudorf- Bornstein is almost identical to the design of a vessel discovered in extremely rich princely grave of stage С2 in Gommern (CRFB D6, VII-01-3/1, Taf. 2; 42; 43; 61; 85: 2 4; 87; 116: 2; 117; 122;

423 Thin-Walled Glass Beakers with Wide Mouth and Decoration of Cut Ovals (Eggers 223, 224, 226, 228) 123; 126; 129: 2), and therefore could be used to determine their chronology. This vessel differs from the beakers under analysis with thicker walls and spheroid body (fig ), being closer to vessels Е 218 and Е 222. This burial contained varied goods, including three buckets of type Hemmoor (Е 58 and Е 60) (Becker 1993: Abb 4, 6; Becker 2001: 162 Abb.). A chronology of variant Sigerstad-Neyzats of wide beakers with cut decoration The finds of the beakers of this variant in Sigerstad (Е 223), Redlino (Е 226), and Kobbea (Е 228) (fig , 19, 20) were not ac companied with goods capable of clarification of their chronology. 4. Pruszcz Gdanski 7, burial 250 A (fig ). The chronology of this assemblage is determined by the following finds: two buckles (type Varpelev / Keller А and МL В9), crossbow fibula with returned foot А 161 / Gorokhovskiy А, bone comb of Sigrid Thomas variant 12A, and wheelmade bowl. The buckle has oval frame, tongue with zoomorphic end, and oval panel (fig. 1. 3), the features comparable with the Varpelev buckle and Erwin Keller s provincial Roman type А. In my point of view, the question when these artefacts appeared is directly related with the problem of their origin, which is still not solved, therefore, these buckle are not a reliable chronological indicator. Although in western Roman provinces they emerged in the early fourth century, we have every reason to suppose that some finds in the Danube provinces were earlier, and the origin of this type was related to the northern Black Sea area. Buckles of type В9 (fig. 1. 4) in Renata Madyda- Legutko s corpus (Madyda-Legutko 1986: 213) are presented by the only specimen from burial 76 in the Przeworsk cemetery of Michalcowo: the publishers date it back to stage С1 (Okulicz, Okulicz 1976: 450, 459). Another buckle of the kind was discovered in Chersonesos. Daniil Kostromichyov has related it to the circle of finds reflecting the presence of Roman garrison in that city (Костромичев 2006: 51 52, 94, рис. 3. 5). Madyda-Legutko has pointed out that type В9 is synchronous with more popular type В7, with the only difference in the shape of the tongue. Types В7 9 date back to the early stage of the Late Roman 1 The finds are numbered according to the catalogue. 423 period (Madyda-Legutko 1986: 13 14). Buckle В7 was discovered in warrior grave Lundtaigen dating from the late phase С1b and most of phase С2 (Bemmann, Hahne 2004: 483, Abb , 484, 307; Ilkjær 1990: 322, Abb. 201; Godłowski 1994: 176, Abb. 5). Therefore, the most probable date of buckles В7 and B9 is phase С1b and the most of phase С2 (except its final). In the Wielbark culture, fibulae А161 / Goro khovskiy А (fig. 1. 1) are typical of phase С1b С2 (Maczynska et al. 2011: 82 83). In the Chernyakhov culture, they appear in assemblages of Oleg Sharov s stages 1 and 2 (AD 220/ /320) (Шаров 1992: chronological indicator 47), though according to Oksana Gopkalalo they are also typical of two earliest chronological groups from the middle to the end of the third century (Гопкало 2008: 96 97, , табл. 3.1, рис. 3. 4). Wheel-made pottery appeared in the Wielbark culture not earlier than stage С2 (Шаров, Бажан 1999: 29). Close bowls with rounded sides and pronounced neck, similar to the find in Pruszcz Gdanski (fig. 1. А7), are known in the burials of cemeteries Leuna and Hassleben (Schulz 1953: e. g. Abb. 33, 34, 71). In the Chernyakhov culture, vessels with pronounced neck (the so-called threepartite ) indicate period 1 (AD 230/40 270/80) according to Ol ga Gey and Igor Bazhan or Sharov s horizons 1 and 2 (AD 220/30 310/20) (Гей, Бажан 1997: 42, chronological indicator 4; Шаров 1992: 187, chronological indicator 40). Therefore, burial 250 А in the cemetery of Pruszcz Gdanski 7 dates from phase С2, possibly except its final years. 5. Bol shaya Bugayovka, burial 100. Fragments of this glass beaker (fig ) were accompanied with fragments of bone comb with concave shoulders and, as the excavator has supposed, trapezium-shaped back (Петраускас, Пастернак 2003: 68). Big width (12 cm) is typical of early combs (Шишкин 2002: 244). According to Gopkalo, combs with trapezium-shaped back are typical of the Chernyakhov chronological groups 2 and 3, from the late third and early fourth century (Гопкало 2008: табл. 3. 1, рис. 3. 4, chronological indicators 44, 56). 6. Neyzats, grave 139. This grave contained two consequent burials. The low burial was ear-

424 424 Aleksandr VASIL YEV Fig. 1. Wide beakers with decoration of cut ovals, and accompanying goods. 1 7 Pruszcz Gdanski 7, burial 250 A; 8 12 Pruszcz Gdanski 7, burial 247; Neyzats, grave 139, burial 2; grave 139, burial 1

425 Thin-Walled Glass Beakers with Wide Mouth and Decoration of Cut Ovals (Eggers 223, 224, 226, 228) lier, accompanied with bronze mirror Khazanov IX (Хазанов 1963), fibula Am15-III/A157, pyramidal bell with extensions on lower corners, bracelet with snake-head terminals Trufanov VI-B (Труфанов 2001: 76), knobbed ring, glass tumbler (fig ), beads, and some other artefacts. The mirror, bracelet, bell and ring are typical of women s grave goods in the Late Scythian culture from ca. AD I do not know similar assemblages, which reliably date back to a later period. As a rule, such sets are accompanied with onepiece bow-shaped fibulae (А156). Two-piece fibula from the grave under analysis is a later chronological form of bow-shaped fibulae, from the mid-third to early fourth century (e. g.: Васильев, Савельев 2008: 32 33). The date of the glass tumbler is the same as of the fibula. In the cemetery of Sovkhoz 10 (Crimea), similar vessels were discovered in graves 20 and 47 with coins of Gordianus III (AD ) and in burial 145 with a coin of Trajan Decius (AD ) (Стржелецкий et al : ). The upper date is marked by the find from vault 9 in the cemetery of Druzhnoye (Crimea) with two different groups of goods. The tumbler of the type of interest, two-piece bow-shaped fibula and plated strap-ends belong to earlier group of finds, from the early fourth century as the latest. The later finds, of the beginning of the fourth century as the earliest, include a series of buckles with solid plain frame and tongue with stepped cut and opposite end bent behind the frame (Храпунов 2002: 67, 179 рис ; 180 рис. 80: 1, 8, 11 15, 34). The second burial in this grave belonged to a child, and was separated by a sub-layer of soil. The beaker was accompanied with a brooch with returned foot and curved bow Gorokhovskiy Б2б, a poorly preserved buckle, and a Bosporan coin minted in AD 313/14 (fig ). Sharov has included fibulae of this variant, as well as brooches Б1, into type 2 (chronological indicator 29) from the Chernyakhov horizons 2 and 3 (AD 280/90 350/60) (Шаров 1992: 183, 186, табл. VIII). The analysis of the distribution of assemblages with fibulae of type 2 according to stages of the cemeteries shows that the specimens of variant Б2 are more typical of horizon 3. According to Gopkalo, fibulae Б2 appeared in graves of group 3 (ca. AD ), were especially popular in the assemblages of the next chronological group (group 4: ca. AD ), and still existed, in a few cases, to the end of the Chernyakhov culture 425 (Гопкало 2008: 97, табл. 3. 1, chronological indicator 52). In my point of view, two asynchronous burials in the same grave, as well as the extra burial of a child, indicate that the chronological gap between them was not too long. Grave goods and funeral rite of the first burial is typical of the so-called Late Scythian culture in the Crimea in ca. AD However, the fibula and the glass tumbler uncover the next chronological stage when the Late Scythian culture did not exist, though we can see that its traditions survived in the second half of the century. Grave goods from the early burial have a wide chronology including the second half of the third to the first years of the fourth century. Taking the consideration about a short chronological gap between the burials into account, one can suppose that the later child s burial was made in the period immediately after the date of the coin discovered in it, though the wider date of the goods from this burial is AD As for the date of the burials, I think that the burial of the woman was made in between of AD , and the child s burial was added to it in AD 313/ Neudorf-Bornstein, burial 4. Rich and varied goods in this grave suggest its attribution to the princely horizon Leuna-Hassleben and its date within phase С2. A special position among the finds accompanying the beaker is occupied by a bronze bucket of type Hemmoor (Е 58) (fig. 2. 5). Another find from the Chernyakhov area, glass beaker from burial 5 in the cemetery of Pereyaslav- Khmel nitskiy, is mentioned in the scholarship as a cut beaker (Rau 1972) or even type Е 223 (Магомедов 2001: 65; Храпунов 2004: 303). However, this attribution of the given vessel is not very convincing because technology is different: the facets were not polished but impressed into hot glass (Сымонович 1977: 180). 8. Pruszcz Gdanski, burial 247. There is a comb and a fibula discovered together with the beaker that allow the one to clarify the chronology of the assemblage (fig ). The comb with high semicircular back and wide out-turned side teeth / legs of Thomas variant I1C (fig. 1. 9). In the Wielbark culture, three-layered combs with semicircular back date back to the latest stage of the Late Roman period (Godłowski 1970: Pl. VII. 31). In the Chernyakhov culture, high back and wide out-turned teeth / legs of analogous combs are considered comparatively late features (Шишкин 2002: ).

426 Fig. 2. Wide beakers with decoration of cut ovals, their analogies, and accompanying goods. 1 4 Haven, burial 4/1971; 5 7 Neudorf-Bornstein, burial 4; 8 9 Neudorf-Bornstein, burial 7; Gommern (5 7, 9 17 not to scale)

427 Thin-Walled Glass Beakers with Wide Mouth and Decoration of Cut Ovals (Eggers 223, 224, 226, 228) Crossbow fibula with returned foot and symmetrical arc-shaped bow (fig ). The body is of solid bar with rectangular cross-section and rich cut decoration. In the Wielbark culture, such classical fibulae А 162 with metope decoration date back to phase С2 (Maczynska 2011: 82 83). In the Chernyakhov culture, similar fibulae, with the only difference in their asymmetric angular bow, belong to Yevgeniy Gorokhovskiy s type Б 1 2, variation г. Similar brooches are most typical of Gopkalo s chronological group 2 from ca. AD (Гопкало 2008: 108). Generally, I date the complex of burial 247 to phase С2, perhaps the late period of it. 9. Skjorringen. As it has already been mentioned, this assemblage dates back to phase С2 by the combination of fibulae Mackeprang IX and III and the composition of necklace (fig. 3. 1, 2, 3, 5). 10. Haven, burial 4/1971. The beaker was accompanied with a fibula А.VI.2 with narrow wedge-shaped foot, a buckle with rectangular twopiece frame, two tiny bronze spear-heads (fig ), a hand-made cup with zoomorphic handle, a hand-made pot, and a wooden bucket set in metal (Schuldt 1973: Аbb ). Keller has dated these fibulae to phase С1b, buckles to С1b / С2, and clay vessels to С1 (Keller 1974: 226, Abb ). Fibulae of the type of interest appear in Ewald Schuldt s Horizon В of the cemetery of Pritzier, which is probably synchronous to Godłowski s phase С2, though in the Elbian circle such fibulae generally are typical of phase С1b (Godłowski 1970: 66, 69 70, Abb 17). In Helga Schach- Dorges relative chronology of the cemetery of Pritzier such fibulae indicate group 2, synchronous to phase С2 (Schach-Dorges 1970: 23, 27, Аbb. 1a, b: 7). Although Madyda-Legutko has attributed the buckle from the grave under analysis to type D30 from phase С2 (Madyda-Legutko 1986: 33 35), I think that the find Haven 4/1971 is closer to type G46, which in the Przeworsk and Western Baltic cultures appeared mostly in phases В2 and В2/С1, though far to the west (including Mecklenburg) was also known in the context of С1b (S. 57). As for the cups with zoomorphic handles, Godłowski and Keller have included them to indicators of phase С1 (Godłowski 1970: Pl. XVI. 3). However, such vessels are also known within the context of early stage С2 (Borby Hansen 2010: 130, Fig. 8). 427 Analogies to tiny bronze spear-heads are well known in graves of the princely horizon Leuna- Hassleben (Шаров 1999: 195). The burial under analysis is perhaps the latest in the group of rich graves discovered in that place, that contained sets of luxury imports including glass bowls with incised decoration (Е 216) and bronze buckets of type Hemmoor (Е 58). These assemblages date back to the turn of phases С1b and С2. Graves in such small princely cemeteries belonged to one or two generations, so burial Haven 4/1971 was not chronologically remote from burials 1/1868 and 7, and we can exclude the end of phase С2 could be excluded from it chronology. 11. Neudorf-Bornstein, burial 7 (fig ). Rich and varied goods in this grave (fibulae А.VI.2, spurs Leuna В, neck-ring, and belt fittings) allow the one to attribute it to the princely horizon Leuna-Hassleben and to date it within phase С2, i. e. ca. AD (Carnap-Bornheim 2002: 15; Авegg-Wigg 2006: 279). In the context of our study, we should also mention several burials with small fragments of cut beakers that do not allow reconstruction of their form and ornamentation and therefore could be attributed to this or that group of vessels only hypothetically. Among these assemblages there are both graves in Varpelev (woman s alpha and man s а) and burial 1/1926 in Leuna. In all the three cases, there were small fragments of vessels of colourless glass (in Varpelev, with tints of green) decorated with oval and circular cuts. Most likely, vessels from these graves belong to series Sigersted. Therefore, our analysis of the chronology of the finds accompanying wide glass beakers of variant Sigersted-Neyzats suggests that these vessels date within phase С2. The time when beakers E 230 appeared It was a certain period within phase C2 when wide beakers of variant Sigersted-Neyzats coexisted with other forms typical of this phase, first of all ribbed bowls Е and variant Widderwill of bowls with ring-foot. Large beakers of type Kowalk (Е 230) are the most typical form in phase C3 to replace all the variants of wide beakers with cut decoration. However, it is considered that the earliest specimens of this form appeared in Scandinavia as early as phase С2, in Hundstad and Vallstenarum (Straume 1987: Taf , 96).

428 428 Aleksandr VASIL YEV Fig. 3. Wide beakers with decoration of cut ovals, beakers E 230, and accompanying goods. 1 6 Skjorringen; 7 11 Hundstad; Vallstenarum;19 Redlino; 20 Kobbea; 21 Velikaya Bugayovka, burial 100; 22 Sigersted (1 5, 7 10, 12 14, not to scale)

429 Thin-Walled Glass Beakers with Wide Mouth and Decoration of Cut Ovals (Eggers 223, 224, 226, 228) 429 Fig. 4. Distribution map for wide beakers with decoration of cut ovals. 1 Sigersted; 2 Redlino; 3 Kobbea; 4 Pruszcz Gdanski 7, burial 250 A; 5 Velikaya Bygayovka, burial 100; 6 Neyzats, grave 139, burial 2; 7 Neudorf-Bornstein, burial 4; 8 Pruszcz Gdanski 7, burial 247; 9 Skjorringen; 10 Haven, burial 4/1971; 11 Neudorf-Bornstein, burial 7 (all numbered according to the catalogue) In the first case (fig ), the form and ornamental pattern of the beaker (two lines of wide oval cuts above and one line of circular cuts below) and its height makes it similar to wide beakers of series Sigersted. Only proportions differ a little. Rim diameter of the Sigersted-Neyzats beakers is a little bigger than the height of the vessel, though the height of the Hundstad beaker is, on the contrary, 0.4 cm bigger than its maximum diameter. Its proportions are almost identical with the Neyzats beaker: the height of the latter is 0.3 cm bigger than its mouth diameter. Fibulae Maketprang IX and II, spiral wire beads, and a rich set of polychrome beads (fig ) of the Hundstad assemblage are similar to the mentioned above rich woman s grave in Maglebjerg from the early phase С2 (Borby Hansen 2010: 193; on the chronology of Scandinavian necklaces see: Boye 2012; Lund- Hansen 2012). As for the Vallstenarum beaker (fig ), its early date is based mainly on the analogy to solid fibula (fig ) from this burial. Straume has referred to the fibulae ornamented with applied wire rings from graves of the group Sakrau-Hassleben (Straume 1987: 116). Sharov has pointed out a more definite analogy, a fibula of Mechthild Schulze s group 23 in the princely grave Straze II (Шаров 1992: ). In my point of view, despite the obvious similarity between many features of these fibulae, the Vallstenarum find is closer to group 47, also originated from Gotland. Schultze has also mentioned the latter analogy. Birger Nerman has dated the fibula in question to the late fourth or early fifth century, though Schulze has suggested AD (Schulze 1977: 39, Tab. 11). There was a belt set, important for the chronology of the assemblage (fig ). Analogous

430 430 Aleksandr VASIL YEV Fig. 5. Ornamentation patterns of glass beakers with decoration of cut ovals. 1 Sigersted; 2 Vallstenarum; 3 Högom; 4 Skjorringen; 5 Druzhnoye; 6 Bremsnes; 7 Himlingøe; 8 Neudorf-Bornstein; Mæle (all not to scale) fittings appeared in warrior graves of Ilkjær s group 10, corresponding to phase С3 (Ilkjær 1990: 302, 305, 307, Аbb. 193; Godłowski 1994: Abb. 1). In my point of view, this chronology is most probable for the entire assemblage. Therefore, the Hundstad beaker dates back to phase С2, but its morphological features are, in my point of view, closer to wide beakers of variant Sigersted rather than to beakers Е230, and the Vallstenarum beaker dates back to phase С3, similarly with other classical specimens of type Е230. Conclusion The analysis undertaken has shown that variant Sigerstad-Neyzats of thin-walled glass beakers with wide mouth and decoration of cut ovals occured in the area between the Baltic and the Black Sea in phase С2. I can draw preliminary conclusion that the most numerous series Sigersted probably existed during the entire phase, though series Haven occurred, according to the materials in possession, in the early period, and series Skjorringen in the late period of the phase. All the three series have the ornamental patterns that also decorated glass beakers from later periods, namely phases C3 and D, in their entire distribution area. Therefore, we have not enough arguments to suggest narrow areas where individual series existed. The impression is that the height and proportions of the vessel form a chronological indicator. In this regard, demonstrative is the comparison of beakers from Neyzats and Hundstad. The height of both vessels is a little big bigger than the mouth diameter (0.3 and 0.4 cm respectively). The height of these beakers put them to specific positions among their types: the Neyzats beaker is 13.5 cm high, more likely to type Е230, though the Hundstad

431 Thin-Walled Glass Beakers with Wide Mouth and Decoration of Cut Ovals (Eggers 223, 224, 226, 228) 431 beaker is 8.9 cm high, similarly to beakers Е223. Therefore, the Neyzats beaker accompanied with a coin from the early fourth century is the latest among the vessels of its type, and the Hundstad beaker is the earliest of its type. In phase С3, wide beak ers of variant Sigerstad- Neyzats (Е 223, 224, 226, 228) were replaced with elongated cylindrical beakers of variant Kowalk (Е230) almost simultaneously in the entire area in between of the Baltic and Black Sea. Catalogue 1. Sigersted (Е 223), Zealand island, Denmark. Colourless glass with tints of green. Height: 11 cm, diameter: 13 cm. Walls cm thick. Two lines of oval and one line of bigger cuts (Eggers 1951: no. 205; Rau 1972: 120; Straume 1987: no. 89.1). 2. Redlino (Е 226). The Dębczyn group, Poland. Colourless glass with tints of green. Height: 7 cm, diameter: 8.6 cm. A line of oval and a line of circular cuts (Eggers 1951: no. 702; Rau 1972: no. 18). 3. Kobbea (Е 228), Bornholm island, Denmark. Colourless glass with tints of green. Height: 7.7 cm, diameter: 8.4 cm. Walls 0.2 cm wide. A line of oval and a line of circular cuts (Eggers 1951: no. 292; Rau 1972: no. 123; Straume 1987: no. 84). 4. Pruszcz Gdanski 7, burial 250 A. (Е ). Wielbark culture, Poland. Colourless, with tints of green. Нeight: ca. 10 cm. Two lines of oval cuts (Pietrzak, Tuszynska 1988: pl. 373). 5. Velikaya Bugayovka, burial 100 (Е 223). Chernyakhov culture, Ukraine. Colourless glass. Нeight: ca. 9 cm, diameter: 10.5 cm. Walls cm thick. Fragmented. Two lines of oval cuts and one line of narrow horizontal ovals (Петраускас, Пастернак 2003: 67 68). 6. Neyzats, grave 139, burial 2 (Е 223). Crimea, Ukraine. Colourless glass with tints of yellowish green. Нeight: 13.5 cm, diameter: 13.2 cm. Two lines of oval and one line of circular cuts (Храпунов 2004). 7. Neudorf-Bornstein, burial 4. Schleswig-Holstein, Germany. Only top fragment with ornamental band of wide ovals survived. Most likely, there were three bands of the kind, so the vessel belonged to typical form Е 223. Colourless glass (Авegg-Wigg 2008: 33-Аbb. rechts; CRFB D5, XXIV-11-11/1.7). 8. Pruszcz Gdanski 7, burial 247 (Е224). Wielbark culture, Poland. Colourless glass. Нeight: ca. 9 cm. Top fragment is missing. Two lines of narrow oval cuts, each with double line below; the bottom is divided by incised double lines, with horizontal narrow oval in each sector (Pietrzak, Tuszynska 1988: pl. 372) 9. Skjorringen (Е 224), Zealand island, Denmark. Colourless glass with tints of green. Height: 10.6 cm, diameter: 11.2 cm. A line of narrow ovals between two lines of wide cuts, divided by double horizontal lines (Straume 1987: , no. 90, Taf. 103). 10. Haven, burial 4/1971. Mecklenburg-Vorpommern, Germany. Colourless transparent glass. Height: 10 cm, diameter: 13.6 cm. Three lines of narrow oval cuts, chequered, and a line of circular cuts near the bottom, with double horizontal line above and below (Shuldеt 1973; CRFB D3, II-09-6/1.21). 11. Neudorf-Bornstein, burial 7. Schleswig-Holstein, Germany. Colourless glass with tints of bright green. Height: 11.5, diameter: 12.9 cm. Ornamented identically with previous vessel (Авegg-Wigg 2006; 2008; CRFB: D5, XXIV-11-11/1.11). Bibliography Abegg-Wigg А. Zu den Grabinventaren aus den Furstengräbern von Neudorf-Bornstein // Aktuelle Forschungen zu Kriegsbeuteopfern und Furstengräbern im Barbaricum / Schriften des Archäologischen Landesmuseums Ergänzungsreihe. 4. Neumünster, Abegg-Wigg А. Glas der Fursten. Zu den reichen Grabern von Neudorf-Bornstein aus der Romischen Kaiserzeit // Archaologische Nachrichten aus Schleswig-Holstein. 14. Neumünster, Becker M. Die Römischen Fundstucke aus dem germanischen Furstengrab der spatrömischen Kaiserzeit bei Gommern Lkr. Burg // Germania /2. Becker M. Luxuriöser Hanshalt fur den Toten // Gold fur die Ewigkeit. Das germanische Furstengrab von Gommern. Halle (Saale), 2001.

432 432 Aleksandr VASIL YEV Bemmann J., Hahne G. Waffenführende Grabinvetare der jüngeren römischen Kaiserzeit und Volkerwanderungszeit in Skandinavien. Studie zur zeitlichen Ordnung anhand der norwegischen Funde // Bericht der Römisch- Germanischen Komission Borby Hansen B. Kvinden på Maglebjorg. En rig grav fra yngre romersk kjernalder ved Næstved // Aarbøger for nordisk oldkyndieched og historie København, Boye L. The Rich Graves from Late Roman Iron Age on Estern Zeeland with a Special View on Glass Beads // Inter Ambo Maria. Northern Barbarians from Scandinavia towards the Black Sea (Abstracts). Simferopol, Carnap-Bornheim C. von. Zu den Prachtgürteln aus Neudorf-Bornstein (Kr. Rendsburg Eckernförde) Grab 3 und Grab 7 // Drik og du vil leve skønt. Festkrift til U. Lund Hansen på 60-arsdagen 18. augst Copenhagen, Corpus der romischen Funde im Barbaricum. Deutschland. 3. Bundeslsnd Mecklenburg-Vorpommern. Bonn, Corpus der romischen Funde im Barbaricum. Deutschland. 5. Hamburg, Schleswig-Holstein. Bonn, Corpus der romischen Funde im Barbaricum. Deutschland. 6. Sachsen-Anhalt. Bonn, Eggers H.-J. Der römische Import im freien Germanien. Hamburg, Eggers H.-J. Zur absoluten Chronologie der römischen Kaiserzeit im freien Germanie // Jahrbuch des RGZM Godłowski K. The Chronology of the Late Roman and Early Migration Periods in Central Europe / ZNUJ PA. 11. Krakow, Godłowski K. Die Chronologie der germanishen Waffengreber in der jungeren und spaten Kaiserzeit // Beitrage zu romischer und barbarischer Bewaffnung in den ersten vier nachchristlichen Jahrhunderten. Lublin; Marburg, 1994 Ilkjær J. Illerup Ådal Die Lanzen und Speere / Jutland Archeological Society Publications. XXV/1 2. Højbjerg,1990. Keller E. Zur Chronologie der jüngerkaiserzeitlichen Grabfunde aus Südwestdeutchland und Nordbayern // Studien zur vor- und frugeschichtlichen Archäologie. Festschrift für J. Werner zum 65 Geburdstag. Munchen, Lund Hansen U. Römischer Import im Norden // Nordiske Fortidsminder Lund-Hansen U. What about Beads? Glass and Amber Beads in the Late Roman Iron Age Relation between Scandinavia and the Black Sea? Thoughts about Production and Trade // Inter Ambo Maria. Northern Barbarians from Scandinavia towards the Black Sea (Abstracts). Simferopol, Maczynska M., Andrzejowski J., Gorecki J., Hammer P., Pawlikowski M., Rudnicka D., Voß H.-U. Der frühvölkerwanderungszeitliche Hortfund aus Lubiana, Kreis Koscierzyna [Pommern] // Bericht der Römisch-Germanischen Kommission. 90. Mainz am Rhein, Madyda-Legutko R. Die Gürtelschnallen der Römischen Kaiserzeit und der frühen Völkerwanderungszeit im mitteleuropäischen Barbaricum // BAR IS Okulicz Ł., Okulicz J. Cmentarzyska kultury pomorskiej i z okresu rzymskiego w Michalkowie, gm. Dobzyn, woj. Wloclawek // Wiadomosci Archeologiczne XLI. Pietrzak M., Tuszynska M. Periode romaine tardive (Pruszcz Gdanski 7) // Inventaria Archaeologica. LX. Warszawa; Lodz, Rau G. Korpergraber mit Glasbeigaben des 4. nachchristlichen Jahrhunderts im Oder-Weichsel-Raum // APA Rau G. Spatantike Facettschliffglässer im Nord- und Osteuropa // APA Schnach-Dorges H. Die Bodenfunde des 3 bis 6 Jahrh. n. Chr. Zwischen unteren Elbe und Oder // Offa-bucher Schulz W. Leuna, ein germanischer Bestattungsplatz der spätrömischen Kaiserzeit / Deutsche Akademie der Wissenschaften zu Berlin. Schriften der Sektion für Vor- und Frühgeschichte. 1. Berlin, Schulze M. Die spatkaiserzeitlichen Armbrustfibeln mit festen Nadelhalter (Gruppe Almgren VI, 2) / Antiquetas. Abhandlungen zur Vor- und Frühgeschichte, zur klassischen und provinzial-römischen Archäologie und zur Geschichte der Altertumbs. 17. Bonn, Shuldt E. Sondierungen auf dem Friedhof der spatromischen Kaiserzeit von Haven, kreis Sternberg // Bodendenkmalpflege in Mecklenburg. Jahrbuch Berlin, 1973.

433 Thin-Walled Glass Beakers with Wide Mouth and Decoration of Cut Ovals (Eggers 223, 224, 226, 228) 433 Straume E. Glässer mit Facettenschliff aus skandinavischen Gräbern des 4. und 5. Jarhunderts n. Chr. // Universitetsforlaget. Ser. B: Skrifter. LXXIII. Oslo, Straume E. Fibeln der römischen Kaiserzeit aus Norwegen der Stand der Forschung // 100 Jahre Fibelformen nach Oscar Almgren. Forschungen zur Archäologie im Land Brandenburg. 5. Wunsdorf, Анисимов А. И. О продвижении племен готского союза в Северо-Восточное Причерноморье в середине III в. н. э. (по нумизматическим данным) // Скифия и Боспор. (Материалы конференции) Новочеркасск, Васильев А. А., Савельев О. К. Переход от начального к финальному этапу позднесарматской культуры в междуречье Днестра и Дуная (хронологический аспект) // Germania Sarmatia. Древности Центральной и восточной Европы эпохи римского влияния и переселения народов. Калининград, Гавритухин И. О. Хронологические индикаторы финала черняховской культуры. 1. Толстостенные кубки со шлифованной и пластической орнаментацией: конические и с выделенной ножкой (Eggers , Straume VI IX) // Сто лет черняховской культуре. К., Гей О. А., Бажан И. А. Хронология эпохи «готских походов». На территории Восточной Европы и Кавказа. М., Гопкало О. В. Бусы и подвески черняховской культуры. К., Гороховский Е. Л. Хронология черняховских могильников Лесостепной Украины // Труды V международного конгресса археологов-славистов. 4. К, Костромичёв Д. А. Римское военное снаряжение из Херсонеса // МАИЭТ ХII/1. Магомедов Б. В. Черняховская культура. Проблема этноса / Monumenta Studia Gothica. I. Lublin, Петраускас О. В., Пастернак В. В. Скляні посудини могильника черняхівської культури Велика Бугаївка у Середньому Подніпров ї // Археологія Стржелецкий С. Ф., Высотская Т. Н., Рыжова Л. А., Жесткова Г. И. Население округи Херсонеса в первой половине I тысячелетия новой эры // Stratum plus Сымонович Э. А. Стеклянная посуда из поднепровско-причерноморских памятников черняховской культуры // СА Труфанов А. А. К вопросу о хронологии браслетов с зооморфными окончаниями (по материалам крымских могильников) // Поздние скифы Крыма / ТГИМ М., Хазанов А. М. Генезис сарматских бронзовых зеркал // СА Храпунов И. Н. Могильник Дружное (III IV вв. н. э.) / Monumenta Studia Gothica. II. Lublin, Храпунов И. Н. Подбойная могила позднеримского времени из некрополя Нейзац // ХСб ХIII. Храпунов И. Н. Некоторые итоги исследования могильника Нейзац // Исследования могильника Нейзац. Симферополь, Шаров О. В. Хронология могильников Ружичанка, Косаново, Данчены и проблема датировки черняховской керамики // ПАВ Шаров О. В. О времени появления Т-образных шарнирных фибул в Северном Причерноморье // Stratum plus Шаров О. В., Бажан И. А. Черняховские этюды // Stratum plus Шишкин Р. Г. Хронологические признаки трехслойных гребней черняховской культуры // Сучасні проблеми археологіï. Київ, 2002.

434 434 Aleksandr VASIL YEV Александр ВАСИЛЬЕВ Тонкостенные стеклянные кубки с широким устьем, украшенные шлифованными овалами (Еggers 223, 224, 226, 228) Резюме Шлифованный декор из различных комбина ций округлых фасеток и врезных линий был характерен для нескольких групп стеклянных сосудов позднеримского времени, получивших широкое распространение между Балтикой и Чёрным морем. В первую очередь, это полусферические чаши Е 216 и цилиндрические кубки Е 230. Промежуточное положение между ними занимает гетерогенная группа стеклянных сосудов, относящаяся к типам Е 220, 221, Как правило, максимальный диаметр этих сосудов незначительно пре восходит их высоту. С формальной точки зрения это позволяет оп ределить данную группу стеклянных изделий как «чаши», однако в большинстве публикаций они фигурируют как «кубки», что подчёркивает их большее сходство с сосудами типа Е 230, чем с типом Е216. Поскольку именно пропорции отделяют их от прочих тонкостенных стеклянных кубков с аналогичным декором, предлагается именовать их «кубками с широким устьем» или, для краткости, «широкими кубками». По особенностям декора и некоторым другим признакам они были разделены на четыре варианта. Предметом данного исследования яв ля ются кубки орнаментированные комбина циями из овалов различного размера и го ризон тальных врезных линий выделенные в вариант «Sigerstad-Нейзац» (Е 223, 224, 226, 228). Внутри варианта выделяется три серии кубков по различиям в схеме орнаментации: серия 1 («Si gerstad») два или три ряда ши роких овалов и окружности; серия 2 («Skjor ringen») че редующиеся ряды узких и ши роких овалов, сдво енные горизонтальные ли нии; серия 3 («Hä ven») три ряда узких овалов и ряд окружностей в придонной части. Анализ хронологии комплексов с рас сматри ваемыми сосудами показал, что они да тиру ются в рамках этапа С2. Вероятно, самая мно гочисленная серия 1 существовала на протяжении всей фазы, серия 2 появилась на её ран нем отрезке, серия 3 на позднем. Все три схе мы декора прослеживаются в оформлении стек лянных кубков этапов С3 и D. На этапе С3 прак тически одновременно на пространстве меж ду Балтикой и Чёрным морем на смену кубкам варианта «Sigerstad-Нейзац» приходят кубки варианта «Kowalk» (Е230).

435 435 Marcin WOŹNIAK Some Remarks on Settlement Structures of the Wielbark Culture in the End of the Roman Period and Early Migration Period in Middle-Eastern Poland (Masovia and Podlachia) The problem of settlement transformations of the Wielbark culture is one of more interesting issues of the archaeology of the Central European Barbaricum. From its crystallisation till the end of its existence, this unit was very dynamic, which manifested itself in, among other things, considerable territorial changes of settlement recorded in this area (Wołągiewicz 1981a: 154; 1981b: 82 87). Particular attention has been paid to the formation process of this culture as well as to the migrations of some settlers from Pomerania to the South (cf. Mączyńska 2007, with primary literature). Both these facts recorded in ancient sources are associated with the migration of the Gothic tribes from Scandza island to the mythical region of Oium. They were one of the sources of significant cultural transformations in Central Europe related to e. g. the development of the Chernyakhov culture (Kokowski 1988: 17 21). The increase in source materials which is recorded recently makes it possible to attempt at learning about some less known aspects of the Wielbark culture settlement, not accounted for in ancient sources. One of them is the issue of the decline of the Wielbark culture settlement in eastern Masovia and Podlachia as well as western Belarus. This is one of the least investigated issues of the Late Antiquity. But it is important because at that time, one of the main routes connecting the Black Sea coast with Scandinavia ran through this region. The Wielbark Culture settlement spread to the region in question in the middle of the second century AD (phase B2/C1). The new settlers came to the previous domains of the Przeworsk culture people, probably assimilating a part of the local community, and occupied territories beyond its compact range (Andrzejowski 2001a: ). The Wielbark culture settlement was in full bloom in the second half of the third and early fourth century AD (phase C1b C2) (fig. 1A). It included most of areas located in the drainage basins of major rivers such as the Narew and the Bug (Masovian and Podlachian Lowlands); its western border was the line of the river Vistula and the river Skrwa, the northern border was the marshy areas on the outskirts of Masuria, the eastern border was the area of the present south-western outskirts of Belarus, and the southern border was the axis of the river Wieprz (Andrzejowski 2007: , with earlier literature). The latest finds from this area which can be associated with the Wielbark culture date from the early phase of the Migration period. In the literature, it has been noticed that their number decreased considerably in comparison with the previous period (Godłowski 1985: 122; 1989: 23 37; Andrzejowski 2001: ), but it has also been pointed out that all Masovian-Podlachian settlement clusters were continuously used from the advanced stage of the Late Roman period to the early stage of the Migration period (Andrzejowski 2001a, 112). While the study of maps of settlement points dated with certainty to phases C1b C2 and C3 D seems to confirm these observations, the analysis of the chronological structure of materials obtained at individual cemeteries, and more broadly within settlement zones, indicates significant differences between some areas within the Masovian-Podlachian province of the Wielbark culture. Coming to the main subject of this text, one should pay attention to the basic problem hindering the study of the Wielbark culture settlement in Masovia and Podlachia, namely the insufficient state of research on the chronology of this cultural unit in this area (cf. Andrzejowski 2007: ,

436 436 Marcin WOŹNIAK

437 Some Remarks on Settlement Structures of the Wielbark Culture in the End of the Roman Period and Early Migration Period in Middle-Eastern Poland Fig. 1. A Territory of the Wielbark Culture in the Late Roman period (the province of the Wielbark Culture in the Masovia and Podlachia was marked by a dark tone). B Settlement of the Wielbark Culture in the Masovia and Podlachia. The sites inhabited till phase C1b C2 (black dots) and phase C3 D (red points). The sites mentioned in text are labeled by numbers: 1 Brulino-Koski, district Ostrów Mazowiecka; 2 Cecele, district Siemiatycze; 3 Goździk, district Garwolin; 4 Jartypory, district Węgrów; 5 Kleszewo, district Pułtusk; 6 Kłoczew, district Ryki; 7 Kołoząb, district Płońsk; 8 Kozłówko, district Nidzica; 9 Krupice, district Siemiatycze; 10 Kutowa, district Hajnówka; 11 Litwinki, district Nidzica; 12 Modła, district Mława; 13 Nadkole, district Węgrów; 14 Niedanowo, district Nidzica; 15 Oronne, district Garwolin; 16 Stara Wieś-Ludwinowo, district Węgrów; 17 Stupsk, district Mława (Andrzejowski, Cieśliński 2007, with author s modifications) 437 with earlier literature). Due to the fact that many sites were examined only to a small extent and to the lack of studies of materials from large cemeteries, most findings were based on a relatively small group of artefacts, which obviously influenced the conclusions. The situation is even more complicated because of the peculiar funeral rite of the local community, which is characterised by a much more modest (in terms of quality) and less numerous set of artefacts placed in grave pits. Due to both these elements it had not been possible to develop a precise system of relative chronology of the local materials. Moreover, their specificity makes it impossible to use the detailed chronological systems developed for the materials of the Masłomęcz group (cf. Kokowski 1997: ) or for some Pomeranian cemeteries (e. g. Weklice, district Elbląg, see Natuniewicz-Sekuła, Okulicz- Kozaryn 2011: ). As a result, our view of the Wielbark culture in Masovia and Podlachia is static. In the light of the findings to date, the most interesting, final phase of the Wielbark culture settlement in Masovia and Podlachia is relatively well observable in the archaeological material. It is mainly characterised by various forms of crossbow brooches with a solid catch plate A VI, 2, e. g.: caterpillar brooches, with plates on the bow, with a knob on the head; massive buckles with an oval or round thickened frame; flat fittings of the end of the belt and bone or antler combs with bell-shaped grips (fig. 2). Unfortunately, among these forms it has not been possible to isolate materials that can be unambiguously associated with the earlier or the later period within this stage. However, the mass increase in source materials from Masovia and Podlachia recorded in the last 20 years as well as new detailed studies of selected groups of artefacts make it possible to re-examine the chronology of selected finds in the local environment. The diagram of co-occurrence of well dated forms from phases C1b C2 and C3 D shows that some artefacts which have been associated with phases C2 and C3 D so far cannot be dated within such a broad chronological scope (fig. 3). It concerns mainly the group of brooches with a returned foot and a plate pin rest, similar to type 1 A. 172; almost all items of later forms with a solid catch-plate; fibulae with a frame-like foot of type A. 185 and similar forms; glass beakers 1 The classifications of artefacts are based on: brooches Almgren 1923; buckles Madyda-Legutko 1987; Roman imports Eggers 1951.

438 438 Marcin WOŹNIAK

439 Some Remarks on Settlement Structures of the Wielbark Culture in the End of the Roman Period and Early Migration Period in Middle-Eastern Poland with cut ornament E 230 and oval buckles with a thickened frame of types H (fig. 4a-c; 5a; 6ab). These items do not co-occur with the indicators of phase C1b C2, such as brooches of types A , 170, buckles of types D or the latest Roman industrial imports produced mostly in Rhineland workshops, brought within the so-called Danish wave (cf. Wołągiewicz 1970: , diagram A), however, are found with artefacts dated with certainty to phase C3 D. An exception to this rule are only the classic fibulae with a returned foot and a knee-shaped bow (similar to variants FM 71, FM 103, FG 79 Kokowski 1997: ), which occur mainly in phase C2 (Kokowski 1997: ; Cieśliński 2010: 62). This data clearly indicates a late chronological position of the group in question and makes it possible to associate it rather with phase C3 D. Probably a considerable number of them should be associated with the earlier part of this period, overlapping to some extent with the end of the previous phase, which is indicated by the above-mentioned brooches with a returned foot. Thanks to these findings, it is possible to reexamine the chronological structure of cemeteries of individual settlement concentrations. They also make it possible to modify the map of settlement points of the Wielbark culture from phase C 3 -D in Masovia and Podlachia. Taking into account the above data concerning the chronology of different forms of artefacts, so far around 60 sites have been recorded in the Masovian-Podlachian province of the Wielbark culture, where materials that can be associated with the end of the Roman period and the beginning of the Migration period have been found (fig. 1B). Their distribution indicates a clear concentration of settlement in northern Masovia: in the territories along the drainage basins of the Wkra, the Sona and the Łydynia and in areas located on the lower Narew, near the mouths of its 439 tributaries the Orz and the Orzyc. Relatively numerous settlement points have also been recorded in Podlachia and western Belarus, in the interfluve of the Narew and the Bug, where a few local settlement concentrations can be distinguished. The number of sites from phase C3 D here can increase slightly if one accepts the thesis about the late chronology of barrows covering cremation graves with modest grave goods, which are dated to this period only on the basis of the rite (Jaskanis 2012, passim). Smaller but clear concentrations of sites with materials from this phase are present along the Liwiec and in the bifurcation of the Vistula and the Narew (near Warszawa). Other isolated points scattered across the entire Masovian-Podlachian province of the Wielbark culture mark the maximum extent of its settlement at the close of the Roman period and at the beginning of the Migration period. What is noteworthy is a clear strip of settlement gaps stretching from the middle course of the Vistula across the South Podlachian Lowland and West Polesia. A comparison of dispersion of points dating from phase C3 D with a map of settlement of the Wielbark culture in phase C1b C2 indicates visible differences in settlement intensity of individual areas. The smallest divergences are recorded in northern Masovia, where the number of settlement points remained more or less the same. In Podlachia, settlements were used continuously till the beginnings of the Migration period as well, but the number of points from this period which are dated with certainty is smaller. The situation is similar in the case of the strip of sites stretching along the Liwiec and the concentration near the place where the Narew flows into the Vistula. The most distinct changes are recorded along the eastern bank of the Vistula, to the south of Warszawa. Only at one out of several sites from phase C1b C2 recorded there, including the cemetery in Kłoczew, district Ryki Fig. 2. Chronological indicators of phase C3 D in the eastern Masovia and Podlachia. a Kozłówko, district Nidzica, grave XVI; b Tomasze, district Ostrołęka; c Gozdowo, district Sierpc, grave; d Niedanowo, district Nidzica, grave 132; e Cecele, district Siemiatycze, grave 407; f Niedanowo, district Nidzica, grave 364; g Modła, district Mława, grave 177; h Brudnice, district Żuromin, grave; j Kozłówko, district Nidzica, grave VI. Sources: a Woźniak 2011; b Prochowicz 2001; c Musianowicz 1949; d, f, i Ziemlińska-Odojowa 1999; e Jaskanis 1996; g Andrzejowski 2006; h Szela 2010; j Heydeck 1892

440 440 Marcin WOŹNIAK with around 80 graves (Balke 1971) and the cemetery in Goździk, district Garwolin with almost 30 features (Niewęgłowski 1984: ), artefacts that can be associated with phase C3 D have been found. These materials were found at the settlement and the cemetery near the village of Oronne, district Garwolin (Czarnecka 2012). At the first site, a brooch with a solid catch plate and a frame-like foot type A. 185 was found in the cultural layer (fig. 7a). In the Central European Barbaricum, brooches of this form are dated to phases C2 and C3 (Schuster 2004: 86 88), but in the Masovian- Podlachian materials, it is clear that their chronology is late and covers phase C3 D (fig. 3). At the cemetery, an iron buckle with a thickened frame of type H. 15 was recorded (fig. 7b). The presence of materials from the period in question only in one settlement complex may indicate a settlement crisis in this area at the close of the Roman period and at the beginning of the Migration period. However, one should bear in mind that it might be caused by the lack of advanced research in this region. Further studies of the local community settlement at the close of the Roman period and at the beginning of the Migration period should be based on the analysis of the chronological structure of individual cemeteries (fig. 8). Thanks to such an analysis, it is possible to observe the changes of development or decline of the settlement of the community using a given cemetery. The overview of data resulting from the analysis of individual sites located within the same concentrations or settlement areas, supplemented with the conclusions drawn from the analysis of a general map of settlement points, makes it possible to attempt at describing the nature of the local settlement. However, one should bear in mind that such an attempt is very inaccurate due to the insufficient state of research and to our interpretation of archaeological sources. Therefore, it probably does not correspond fully to the actual situation from more than 1,500 years ago. Studies on this issue should be based on a series of large but not necessarily long-lasting sites. Unfortunately, studies of some of them have not been carried out or fully published, which limits the possibility of drawing conclusions. Among the biggest cemeteries, the one that is of fundamental importance for the Wielbark culture in the discussed area is the cemetery in Niedanowo, district Nidzica (Ziemlińska-Odojowa 1999), located in the northern outskirts of Masovia. It provided two hundred five complexes of the Wielbark culture 2 from phase B2/C1 to phase C3 D, that is from the whole period of the presence of the Wielbark culture community in this area. One hundred five of them are features with modest or no grave goods, which cannot be analysed chronologically at all. A clear horizon including thirty six features is represented by the earliest burials from phase B2/C1 C1a, dated with certainty on the basis of e. g. brooches A. 41, 96, 128, fibulae with a high catch plate A. VII, 1; snake bracelets or the urn form of the grave. Much less clear is the horizon from phase C1b C2, associated with thirteen graves including e. g. brooches A. 161/162, 167, 168, 170 or buckles with a D-shaped frame corresponding to Types D Also some of fifteen graves dated broadly to phases C1b D probably date from this phase. There are more graves that can be associated with phase C3 D, namely thirty one. They are characterised by numerous brooches with a solid catch plate A.VI,2; late forms of fibulae with a returned foot, including items with a plate pin rest; glass vessels decorated with cut, pottery decorated with spring imprints. A similar situation is recorded at the nearby cemetery in Kozłówko, district Nidzica (former Klein Koslau, Kreis Neidenburg) (Heydeck 1892; Woźniak 2011). At this cemetery, studied in the nineteenth century, around eighty grave pit features were discovered, and only thirty nine of them had any grave goods. This cemetery was founded in phase B2/C1 C1; but the number of graves from this period is very small. The rest of features which are dated with certainty and which contained similar grave goods to those recorded at Niedanowo date from phases C1b C2 (four graves) and C3 D (ten graves). The prevalence of graves from the latest stage may be even higher, as seven features with pottery shards and glass vessel remains only were not included; these artefacts occur at Kozłówko mainly in graves from phase C3 D. Unfortunately, complete studies of another two important cemeteries of the Wielbark culture have not been carried out or published. The first one was discovered in Modła, district Mława (Grzymkowski 1986; Andrzejowski 2006). Around sixty graves of the Wielbark cul- 2 In the Late Pre-Roman period and the Early Roman period, the cemetery in Niedanowo was used by the Przeworsk culture people; this applies to most of the sites discussed in this text. For the sake of clarity, all values concerning the number of graves are given only in the context of the Wielbark culture burials.

441 Some Remarks on Settlement Structures of the Wielbark Culture in the End of the Roman Period and Early Migration Period in Middle-Eastern Poland 441 Fig. 3. Co-occurrence of selected artifacts in graves of the Wielbark culture from the Masovia and Podlachia (without phase B2/C1a C1a)

442 442 Marcin WOŹNIAK Fig. 4. Kleszewo, district Pułtusk. Grave 884 (part of grave furnishing) (drawing: P. Holub)

443 Some Remarks on Settlement Structures of the Wielbark Culture in the End of the Roman Period and Early Migration Period in Middle-Eastern Poland 443 Fig. 5. Niedanowo, district Nidzica. Grave 539 (Ziemlińska-Odojowa 1999) ture, dated to the period from phase C1 to phase C3 D, were discovered at this site. Among grave complexes with a precisely determined chronology, the most numerous ones are those from the latest phase of cemetery usage, including e. g. brooches A. 172 and A. VI. 2; glass vessels ornamented with cut, buckles with a thickened frame and flat fittings of the end of the belt. The other mentioned site is the cemetery in Kleszewo, district Pułtusk 3. Preliminary research shows that the Wielbark culture cemetery was founded only in phase C1b C2; there is no material that could be associated with phase B2/C1 C1a. The earliest horizon is marked by brooches with a returned foot and a knee-shaped bow, brooches A. 168 or two-part buckles with a D-shaped or omega-shaped frame. The latest horizon of the cemetery, which is very clearly visible, is dated to phase C3 D on the basis of fibulae A. 172, brooches with a solid catch plate and buckles with a thickened frame. According to a preliminary assessment based only on grave complexes dated with certainty on the basis of metal artefacts, the 3 Materials currently under study by the present author.

444 444 Marcin WOŹNIAK Fig. 6. Modła, district Mława. Grave 177 (Andrzejowski 2006)

445 Some Remarks on Settlement Structures of the Wielbark Culture in the End of the Roman Period and Early Migration Period in Middle-Eastern Poland 445 Fig. 7. Oronne, district Garwolin. Settlement, stray find (a); cemetery, stray find (b) (Czarnecka 2012) number of burials from phase C1b C2 is similar to the number of those from phase C3 D. Materials from phase C3 D were also found in Stupsk, district Mława (Grzymkowski 1996: 168, ). From among a dozen or so Wielbark culture complexes, all the features dated with certainty are associated with phases C 2 D (Grzymkowski 1996: 168). The chronological structure of a large cem- etery in Kołoząb, district Płońsk (Tomaszewska 1988; Żórawska 2006) is completely different. At this site, only one feature (grave 321) that could be associated with this phase was found. The feature is dated on the basis of an Elbian brooch with a solid catch-plate and a foot widened at the end. 4 There is a prevalence of graves from the beginnings of the Late Roman period. So far, there are Fig. 8. Chronological structure of selected cemeteries of the Wielbark culture in the Masovia and Podlachia 4 I would like to thank Jacek Andrzejowski of the State Archaeological Museum in Warszawa for the information on the details of the artefact.