AMPHORA RESEARCH

IN CASTRUM VILLA ON BRIJUNI ISLAND

EDITED BY
TAMÁS BEZECZKY

BEZECZKY_Text.indd 1 18.03.19 14:39

 ÖSTERREICHISCHE AKADEMIE DER WISSENSCHAFTEN
PHILOSOPHISCH-HISTORISCHE K LASSE
DENKSCHRIFTEN, 509. BAND

ARCHÄOLOGISCHE FORSCHUNGEN
Band 29

BEZECZKY_Text.indd 2 18.03.19 14:39

 AMPHORA RESEARCH
IN CASTRUM VILLA ON BRIJUNI ISLAND

CONTRIBUTORS

TAMÁS BEZECZKY, PIERO BERNI MILLET, MICHEL BONIFAY,

CLAUDIO CAPELLI, HORACIO GONZÁLEZ CESTEROS,
SÁNDOR JÓZSA, MARTINO LA TORRE, ALEXANDER SCHOBERT,
GYÖRGY SZAKMÁNY

EDITED BY
TAMÁS BEZECZKY

BEZECZKY_Text.indd 3 18.03.19 14:39

 Angenommen durch die Publikationskommission der philosophisch-historischen Klasse der
Österreichischen Akademie der Wissenschaften:
Accepted by the publication committee of the Division of Humanities and Social Sciences
of the Austrian Academy of Sciences by:
Michael Alram, Bert G. Fragner, Andre Gingrich, Hermann Hunger,
Sigrid Jalkotzy-Deger, Renate Pillinger, Franz Rainer, Oliver Jens Schmitt,
Danuta Shanzer, Peter Wiesinger, Waldemar Zacharasiewicz

Redaktion:
Tamás Bezeczky / Katharina Preindl

Umschlagabbildung:
Aerial photograph of Castrum villa (Courtesy of the Brijuni Museum)

Diese Publikation wurde einem anonymen,

internationalen Begutachtungsverfahren unterzogen.
This publication was subject to international and anonymous peer review.
Peer review is an essential part of the Austrian Academy of Sciences Press evaluation
process. Before any book can be accepted for publication, it is assessed by international
specialists and ultimately must be approved by the Austrian Academy of Sciences
Publication Committee.

Die verwendete Papiersorte in dieser Publikation ist DIN EN ISO 9706 zertifiziert und erfüllt die Voraussetzung für
eine dauerhafte Archivierung von schriftlichem Kulturgut.
The paper used in this publication is DIN EN ISO 9706 certified and meets the requirements for permanent archiving
of written cultural property.

Alle Rechte vorbehalten.

ISBN 978-3-7001-7972-6
Copyright © 2019 by
Österreichische Akademie der Wissenschaften, Wien

Satz und Layout: Andrea Sulzgruber, Wien

Druck und Bindung: Wograndl Druck GmbH, Mattersburg
https://epub.oeaw.ac.at/7972-6
https://verlag.oeaw.ac.at
Made in Europe

BEZECZKY_Text.indd 4 18.03.19 14:39

 V

CONTENTS

Foreword (Pülz) .................................. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .  VII

In memoriam Tamás Bezeczky (1949 – 2018) (Krinzinger) .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .  IX
Introduction and acknowledgements (Bezeczky) .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .  XI
List of figures . ....................................... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XIII
List of plates . ........................................ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .  XV

SHORT HISTORY OF THE WORKSHOP AND THE VILLAS

CHAPTER 1 Istria and the amphora workshops (Bezeczky) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 3
1 Historical background .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 3
2 The Fažana workshop .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3 The owners of the workshop .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
4 The amphora stamps .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
5 The production capacity of the villas and the distribution of the Fažana amphorae .. . . .. . . . . . . . . . . . . 7
CHAPTER 2 The villas of Brijuni (La Torre – Bezeczky) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 11
1 Verige Bay (Val Catena) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2 Kolci (Monte Collisi) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 15
3 Castrum . ................. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
CHAPTER 3 The early Christian churches of Brijuni (Schobert) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
1 The Church of St. Mary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 29
2 The early Christian church within the walls of the Castrum and in Verige Bay . . . . . . . . . . . . . . . . . . . . . . . 34
3 The Church of St. Peter .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 34
4 The Bishop of Cissa Pullaria (Cessus) and his residence .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 36

THE AMPHORAE
CHAPTER 4 Italian and Istrian amphorae (Bezeczky) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
1 Italian wine amphorae from the late Republican and early Roman period .. . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
2 Olive oil amphorae . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 42
3 Istrian fish sauces amphorae .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
4 Northern Adriatic amphorae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
5 Miscellaneous Italian amphorae .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 45
Conclusion ................. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 46
CHAPTER 5 Hispanic amphorae (González Cesteros – Berni Millet) .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
1 Iberian amphorae production and the Adriatic region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
2 Identification problems in distinguishing Lusitanian and Baetican amphora types .. . . .. . . . . . . . . . . . . 51
3 The Hispanic amphorae from the Brijuni Castrum villa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Conclusion ................. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 53
CHAPTER 6 Eastern Mediterranean amphorae (González Cesteros) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 55
The eastern Mediterranean amphorae and the north Adriatic region . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 55
The eastern Mediterranean amphorae types from the Brijuni Castrum villa .. . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Conclusion ................. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 69

BEZECZKY_Text.indd 5 20.03.19 19:08

 VI Contents

CHAPTER 7 African amphorae (Bonifay – Capelli) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 71

1 Typology and origin of the African amphorae .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 71
1.1 African amphorae of Punic tradition .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 71
1.2 Roman-African amphorae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 71
1.2.1 Classical Roman-African amphorae .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
1.2.2 Late Roman-African amphorae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 72
1.3 African imitations of non-African types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 73
1.4 Storage jars (?) . .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
2 African imports to Brijuni in the northern Adriatic context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 74
Conclusion ................. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
CHAPTER 8 Amphora stamps and inscriptions (Berni Millet – Bezeczky) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 77
CHAPTER 9 Catalogue (Bezeczky – Berni Millet – Bonifay – Capelli – González Cesteros –
Józsa – Szakmány) . ........ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
PLATES 1-22 .............. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

EPIGRAPHY
CHAPTER 10 Calendar graffiti on Dressel 20 amphorae (Berni Millet) .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
1 The manufacturing stages of a Dressel 20 amphora . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 126
2 Nominal graffiti . . ........ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 126
3 Calendar graffiti ........ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
4 The calendar graffito from the Castrum of Brijuni . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
5 Asiaticus: another paradigmatic case . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 135
6 General reflections ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
7 Updated chart of calendar graffiti in order of the date of the year .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
CHAPTER 11 Inscriptions on laterite finds from Brijuni (Berni Millet) .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
A. Epigraphy on laterite (tegulae, imbrices) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 150
B. Epigraphy on dolia . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
C. General index ........... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
PLATES 23-34 . ........... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173

MICROPETROGRAPHY
CHAPTER 12 Micropetrography of the Fažana amphorae (Szakmány – Józsa) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
Previous works on raw material determination .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 187
Methods and strategy . .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
Results ....................... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
Brief thin section micropetrography of amphorae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
Detailed thin section micropetrography of possible raw materials .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 189
Detailed comparative thin section micropetrography of Fažana amphorae .. . . . . . . . . . . . . . . . . . . . . . . . . . 191
Discussion and conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 194
Summary . .................... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 196
PLATES 35-37 . ........... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197

CHAPTER 13 Summary (Bezeczky) . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201

Sažetak (Bezeczky) .......... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205

APPENDIX The Fažana amphora stamps (Bezeczky) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 209

INDICES .. . . ........................................ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
BIBLIOGRAPHY . . .............................. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 223
CONTRIBUTORS . ............................. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237

BEZECZKY_Text.indd 6 18.03.19 14:39


VORWORT
Die neue Publikation zu den Amphoren aus der Castrum
Villa von Brijuni gibt Anlass zu einigen forschungs­ge­
schichtlichen Reflexionen:
Am Anfang der Forschungen zur Archäologie von Brijuni
stand vor mehr als hundert Jahren Anton Gnirs, Kurator der
altösterreichischen Zentralkommission für Istrien, Rektor
der österreichischen Real-Schule und seit 1909 Direktor
der Staatlichen Antikensammlungen in Pola. Im Auftrag
von Paul Kupelwieser, dem damaligen Eigentümer der Insel,
führte er seit 1904 eine Reihe von Ausgrabungen durch und
legte auch das castrum frei. Er erkannte im Übrigen als Erster
die wesentlichen Funktionen der Ruine, die später von einer
befestigten Siedlung überbaut werden sollte.
Es folgten die Grabung des Museums von Pula durch
die kroatischen Archäologen Štefan Mlakar und Anton
Vitasović mit einigen Präzisierungen zu den Bauphasen
und den landwirtschaftlichen Produktionsabläufen in der
villa rustica.
Außer der im Museum von Pula aufbewahrten Feinkera-
mik wird das gesamte Fundmaterial der Grabungen heute
im Depot des Museums des Nationalparks Brijuni gelagert,
wobei die Amphoren Kollegen Bezeczky zur Publikation
übergeben wurden.
Aus den Fragmenten zahlloser Amphoren, den bewähr-
ten Transport- und Speichergefäßen – sie trugen in der
Regel an den Henkeln eingestempelte Hinweise auf den
Produzenten und Verwalter – ergeben sich für den Fach-
mann faszinierende Einblicke in die ökonomischen Struktu-
ren der römischen Landwirtschaft und in die Handelswege
und kulturellen Beziehungen im gesamten Mittelmeerraum.
Ebendiesen Fragestellungen hat sich Tamás Bezeczky aus-
gehend von seinen 1998 publizierten Forschungen zu den
Laecanius-Amphoren in den letzten Jahrzehnten mit gan-
zer Kraft gewidmet, neue Methoden weiterentwickelt und
dabei wesentliche Resultate erarbeiten können.
Das vorliegende Werk bietet zunächst einen kompakten
Überblick zu allen römischen Villen der Insel, im Speziellen
aber zur Castrum Villa mit ihren zahlreichen Bauphasen.
BEZECZKY_Text.indd 7
VII
Berücksichtigung findet zudem die kleine Kirche, die im spä-
ten 5. Jh. n. Chr. etwas nördlich des castrum errichtet wurde.
Bezeczky führte also – seinem wissenschaftlichen Selbstver-
ständnis entsprechend – die Forschungen nicht alleine durch,
sondern bettete diese in ein internationales Expertenteam
ein, womit er dem Fundmaterial, das chronologisch vom 1.
Jh. v. Chr. bis in das 7. Jh. n. Chr. reicht, und der geographi-
schen Verteilung der Produktions­stätten praktisch im gesam-
ten Kulturkreis des Mittelmeeres gerecht werden konnte.
Ausgangspunkt der vorliegenden Studien waren die Pro-
duktionsstätten in den Latifundien der Familie Laecanius,
die später in den Besitz der Kaiser Vespasian, Domitian
und Trajan übergingen, ehe die Produktion unter Hadrian
eingestellt und durch iberische Güter ersetzt wurde. Daher
sind neben den adriatischen Produktionsstätten vor allem
die iberischen, aber auch die afrikanischen Amphoren von
signifikanter Bedeutung. Zum ersten Mal werden im bislang
oft undifferenziert dargestellten Raum des „Eastern Medi-
terranean“ auch Gruppen von spezifischen Importwaren
bestimmt und einzelnen Orten zugewiesen. Eine große Hilfe
stellte bei diesem Unterfangen die Amphoren-Datenbank
des Autors dar. Für die petrographische Bestimmung der
Materialzusammensetzung der Laecanius-Amphoren haben
auch geologische Untersuchungen zum besseren Verständ-
nis der Herkunft des Tons wesentlich beigetragen.
Die nun posthum vorgelegte letzte Publikation von
Tamás Bezeczky ist ein ganz hervorragendes Beispiel für
seine Arbeitsweise. Das konzise Konzept mit seiner detail-
lierten Zielsetzung wurde in allen wesentlichen Punkten
von ihm bestimmt und gemeinsam mit ausgewählten Fach-
leuten verschiedener Disziplinen umgesetzt. Bezeczky gab
die Gesamtgliederung vor und verfasste die archäologisch-
historischen Überblicke zur Insel und alle Beiträge zu
Laecanius. Auch alle Zusammenfassungen im Werk tragen
seine Handschrift. Dass der Autor, durch seine schwere
Krankheit geschwächt, die Fertigstellung seines bereits zum
Druck eingereichten Werkes, an dem er bis zuletzt gearbeitet
hat, nicht mehr erleben konnte, ist von besonderer Tragik.
18.03.19 14:39
 VIII Vorwort
Es war für das Institut für Kulturgeschichte der Antike
der ÖAW eine selbstverständliche Pflicht, für den geregel-
ten Abschluss des Publikationsprozesses zu sorgen.
Abschließend sei den Autoren gedankt, durch deren
Beiträge das vorliegende Werk zu einer umfassenden wirt-
schafts- und kulturgeschichtlichen Studie wurde. Dan-
kend hervorgehoben seien Horacio González Cesteros
für seine Korrektur­lesungen, besonders des umfangrei-
chen Kataloges, sowie Frau Andrea Sulzgruber für den
Satz und die graphische Gestaltung der Publikation. Zu
danken ist zudem Frau Katharina Preindl für die redaktio-
BEZECZKY_Text.indd 8
nelle Weiterbetreuung des Manuskriptes sowie dem Verlag
der ÖAW und seiner ehemaligen Geschäftsführerin Frau
Mag. L. Triska, die die notwendigen Rahmenbedingun-
gen für die Drucklegung sicher­stellte.
Besonderer Dank gebührt ferner dem Fonds zur Förde-
rung der wissenschaftlichen Forschung (FWF), welcher das
Projekt – wie alle vorangegangenen Forschungen des Autors –
getragen hat, sowie dem Holzhausen-Legat der ÖAW für die
maßgebliche Unterstützung bei der Finanzierung des Drucks.
Wien, 2018 Andreas Pülz
18.03.19 14:39

IN MEMORIAM
TAMÁS BEZECZKY (1949 – 2018)
Da es dem verantwortlichen Herausgeber sowie Hauptautor
Tamás Bezeczky nicht mehr vergönnt war, den Abschluss
und die Publikation des vorliegenden Werkes zu erleben,
soll der Einleitung des Autors eine kurze Würdigung seines
Lebenswerkes vorangestellt werden.
Nach der Ausbildung zum IT-Ingenieur an der TU
Budapest, die er parallel zur Lehrtätigkeit an einer Techni-
ker-Lehranstalt absolvierte, und nach einer mehrjährigen
Verpflichtung am Institut für Konservierung und Methodik
der Museologie wurde Bezeczky im Jahre 1982 Kurator der
Computerabteilung am Ungarischen Nationalmuseum und
inskribierte sich gleichzeitig am Archäologischen Institut
der Universität Budapest. Er stellte ein Forschungsprojekt
zur Bernsteinstraße zusammen und engagierte sich an der
Ungarischen Akademie der Wissenschaften für das Thema
„Computer Application in the Archaeology“. Seine Disser-
tation Roman Amphorae from the Amber Route in Western
Pannonia erschien 1987 in Oxford. Im Vorwort der Dis-
sertation legte Bezeczky seine interdisziplinären Methoden
und archäologischen Ziele programmatisch dar. Die damit
manifeste Doppelbegabung war bestimmend für seinen
weiteren wissenschaftlichen Weg, den er mit konsequenter
Leidenschaft gegangen ist: computergestützte Forschungen
BEZECZKY_Text.indd 9
IX
zu römischen Amphoren mit interdisziplinären Methoden
zur Bestimmung von Material und Produktionsort.
Schon als Student war Bezeczky auf vielen Grabungen
und Fundorten in Ungarn, Jugoslawien und Italien anzu-
treffen. Nach Österreich kam er erstmals im Jahr 1986. Er
war mehrere Jahre als Mitarbeiter bei der Ausgrabung am
Magdalensberg in Kärnten tätig und mit der Bearbeitung
der Amphoren betraut. Ein von der ÖAW in Aussicht
gestelltes Stipendium, welches dann vom Ministerium
leider nicht bewilligt wurde, war neben familiären Über-
legungen einer der Gründe, Anfang der 1990er Jahre von
Budapest nach Wien zu übersiedeln. Die persönlichen
Lebensumstände der Familie waren in diesen Jahren sicher-
lich nicht leicht. Dennoch arbeitete Bezeczky zielstrebig
an der Veröffentlichung seiner Resultate, die grundle-
gende Publikation zu den Amphoren vom Magdalensberg
erschien 1994 in Klagenfurt.
Im Sommer des gleichen Jahres lernte ich Bezeczky im
Grabungsdepot von Carnuntum persönlich kennen und
schätzen. Kiste für Kiste, Regal für Regal durchforstete
er nach „seinen Amphoren“, dokumentierte Formen und
Stempel und untersuchte die Gefäße auf Graffiti sowie auf-
gepinselte Hinweise. Er war für sein erstes Laecanius-Projekt
unterwegs, das von der Soros Foundation (Ungarn) und in
der Folge von der ÖAW und dem ÖAI gemeinsam finan-
ziert wurde. Somit bot sich dem Autor die Möglichkeit zur
vertieften Erforschung der bekannten Produktionsstätten
in Fazana und Brijuni (Istrien), die mit den Latifundien
der senatorischen Familie des Laecanius zusammenhängen.
Die vor Ort erzeugten Amphoren als „Einweggebinde“ sind
Beleg für die im ganzen Mittelmeer verbreiteten landwirt-
schaftlichen Produkte und bieten durch die Stempelreihen
nachvollziehbare Einblicke in die Organisation der Her-
stellung und des Handels. Die viel beachteten Ergebnisse
des Projektes wurden 1998 an der ÖAW publiziert. Mit
den darauffolgenden Aufenthalten als Visiting Fellow am
Department of Archaeology der Universität Southampton
verdichteten sich die internationalen Kontakte Bezeczkys in
besonderer Weise.
18.03.19 14:39
 X In memoriam Tamás Bezeczky (1949 – 2018)
Tamás Bezeczky und Sándor Józsa im September 2012 auf Brijuni
(© G. Szakmány)
Als Grabungsleiter von Ephesos war es für mich nahelie-
gend, den international anerkannten Experten 1998 in das
Forscherteam einzuladen. Die Metropolis Asiae mit dem
großen Handelshafen und dem weltweit berühmten Arte-
mision erwies sich als idealer Ort für die Forschungen zur
Entwicklung und Verbreitung der römischen Amphoren
des östlichen Mittelmeerraumes. In mehr als 10-jähriger
Arbeit schuf Bezeczky in der Tat ein Standardwerk, dessen
Finanzierung dankenswerterweise über den gesamten Zeit-
raum dieses Projektes – nach kompetitiv erfolgten Antrag-
stellungen – vom FWF getragen wurde. Über die traditio-
nellen Bearbeitungsmethoden der Funde hinaus konnten
mit geologischen und petrologischen Methoden mehrere
lokale Produktionsstätten definiert und deren Entwick-
lung über einen größeren Zeitraum dargestellt werden.
Daneben ergab sich für die modernen wirtschaftshis-
torischen Fragestellungen zu antiken Handelsrouten und
Warentransporten eine Fülle von Importkeramik, woraus
eine Vielzahl wirtschaftlicher und kultureller Kontakte
nachgewiesen werden konnte. Von ganz besonderem Inte-
resse war dabei die kontextuelle Betrachtung des gesamten
Fundmaterials aus den bereits abgeschlossenen Grabungen
auf der Tetragonos Agora, sodass auch das chronologische
Gerüst der ausgewiesenen Typen und Varianten abge-
sichert werden konnte. Die Ergebnisse wurden 2013 in
BEZECZKY_Text.indd 10
den Forschungen in Ephesos (FiE XV/1) vorgelegt. In die-
sen Jahren wurde für die übersichtliche Verwaltung aller
Informationen zur großen Zahl der Objekte und für die
systematische Dokumentation der zahlreichen Para­meter
von Bezeczky gemeinsam mit Peter Hornung auch eine
spezielle Datenbank entwickelt. Sie ist in der Einleitung
des Werkes ausführlich beschrieben und konnte in der wei-
teren Folge verdichtet und erweitert werden. Sie ist heute
ein wertvolles Werkzeug für alle Spezialisten der Ampho-
renforschung.
Mit seinem letzten großen Projekt, das wiederum vom
FWF finanziert wurde, kehrte Bezeczky nach Istrien zurück
und setzte es sich zum Ziel, die Amphoren aus einer kleine-
ren Villa des Laecanius (sog. Castrum Villa) an der West-
küste der Insel Brijuni zu bearbeiten. Das Fundmaterial
dieser Villa war den Forschungen Bezeczkys in den 1990er
Jahren nicht zugänglich gewesen. Die schiere Menge der
Fragmente und das breite chronologische Spektrum erfor-
derten eine neue Arbeitsweise.
Wie in der folgenden Einleitung zur Publikation darge-
stellt wird, organisierte Bezeczky ein internationales Team
und brachte das Projekt auf diesem Wege zu einem erfolgrei-
chen Abschluss.
Neben diesen mehrjährigen Forschungen, die jeweils
mit einer Monographie erfolgreich abgeschlossen wurden,
umfasst die Bibliographie Bezeczkys über sechzig Einzel­
titel, Rezensionen und daneben eine Fülle von praxisorien-
tierten Forschungsleistungen, von denen die Datenbanken
zu Ephesos und Brijuni sicherlich die wichtigsten sind.
Tamás Bezeczky ist zu früh von uns gegangen, aber er
hat ein reiches Erbe hinterlassen. Es ist den laufenden und
zukünftigen Forschungen zu den römischen Transport­
amphoren zu wünschen, dass an diesem Erbe weiter gebaut
werden kann.
An den Schluss darf ich mit freundlicher Erlaubnis einen
Satz von Frau Anna Bezeczky stellen:
Ja, mein Mann hatte eine verkürzte Lebensdauer, aber ein
volles Leben! Es wird einem selten zuteil, dass Arbeit, Beruf,
Berufung und Hobby gleichzeitig zutreffen. Er hatte dieses
Phänomen inne, dadurch war er ein glücklicher Mensch!
Wien, 2018 Friedrich Krinzinger
18.03.19 14:39

INTRODUCTION AND ACK NOWLEDGEMENTS
The Austrian Science Fund (FWF) accepted the research
project The “Laecanius amphorae in Brijuni” (P 23684) in
2011. The goal of the project was to publish the new Laeca-
nius amphorae from the Castrum villa in Brijuni Island. The
Laecanius amphora stamps and the villas of Brijuni (Bezec-
zky 1998a) described the amphorae which were available in
the mid-1990s. But at that time it was not possible to study
the amphorae in one of the Laecanius villas (Castrum). The
many thousand finds (amphorae, tegulae, fine ware, glass,
etc.) discovered during the villa excavations, with a few
exceptions, were kept in the archaeological collection of the
Brijuni National Park. Rarely do we have the opportunity to
analyze a group of amphorae from the first century B.C. to
the seventh century A.D. found in such a relatively small
area (1 hectare). This made it clear that all of the excavated
amphora groups needed to be published, not only those pro-
duced by the Laecanius family. An international team was
formed, composed of the specialists who study food produc-
tion in different parts of the Roman Empire and they con-
tributed individual chapters.
In this volume, the reader will find a selection of the
amphorae from the villa. The finds were grouped by place
of production (Adriatic, Iberian, Eastern Mediterranean
and African), and the articles are also published in this
order. Various objects have already been published from
the villa, primarily the architectural elements found dur-
ing the excavations, tiles with stamp (e.g. works by Štefan
Mlakar, Branko Marušić, Vlasta Begović, Ivančica Schrunk,
Anton Vitasović, Robert Matijašić) and ceramic finds (see
papers by Philipp M. Pröttel, Verena Vidrih Perko and Mira
Pavletić). Our team examined all of the currently available
amphora finds and selected the rim, base, neck and handle
pieces of the most important amphora types to be pre-
sented. All of the pieces of the collection which bear a stamp
or inscription have been included. Some 90 amphora types
are represented, which contained olive oil, wine, fruit and
fish products. One must differentiate between locally made
and imported amphorae. The Castrum villa was a place
where olive oil and wine were produced. It is necessary to
BEZECZKY_Text.indd 11
XI
clarify what was produced when. In addition, there are two
detailed analyses. One of them is concerned with a rare graf-
fito on a Baetican amphora, the other with the abundant
tile stamps. The new archaeometrical analyses of the Fažana
amphorae started, they provide new data on the origin and
method of production of the amphorae.
The discovery of the villas on Brijuni Island and the
amphora workshops in Fažana can be attributed to a lucky
accident. In 1893, during the Austro-Hungarian Mon-
archy, a wealthy industrial magnate, Paul Kupelwieser,
bought the Brijuni Islands. He was determined to build a
holiday resort, or in today’s parlance, a wellness and rec-
reation center. With the help of Robert Koch, the famous
microbiologist, he eradicated malaria on the islands. Dur-
ing construction works for the roads and the hotel build-
ing foundations, Kupelwieser’s workers discovered several
Roman ruins. He contracted Anton Gnirs, a teacher in the
“Marinerealschule” of Pula, to lead the archeological exca-
vations. Gnirs discovered several Roman buildings on the
islands, Fažana and in Pula. After the Second World War,
the Yugoslavian president Josip Broz Tito took a liking to
the islands. One of his luxurious homes was in the vicin-
ity of the Castrum villa. Later Tito himself was the driv-
ing force in the excavation of this villa. After his death, the
islands were declared a National Park and parts of it were
opened for the public.
The story of the Castrum villa is closely connected to
the Laecanius family. Ancient and modern authors and the
excavations provide a great deal of information about this
family’s history, economic activity, villas and amphora work-
shop. It is clear that their estates and the income that these
estates produced went to the emperors after the last mem-
ber of the family died without an heir. Less is known about
the amphora workshop and villa in Fažana after the second
century. Although the excavations brought a great deal of
objects to light, our knowledge is incomplete in many ways.
For instance, we do not know who used the villas in the mid-
and late Roman periods. Regarding the excavated walls and
objects, it is unclear which period the published maps belong
18.03.19 14:39
 XII Introduction and acknowledgements (Bezeczky)
to. Various rooms in the villas were altered on several occa-
sions in the Roman period as well as during the reconstruc-
tion following the excavations.
We only provide a short description of the Fažana work-
shop and the villas in Brijuni in this book, as some knowl-
edge of them is necessary to understand the Castrum finds.
There may be some overlap between individual chapters,
but this is unavoidable in order to present the information
coherently. We also refer to the rich bibliography for further
information.
I would like to thank the directors of the Brijuni National
Park, Sandro Dujmović, Eduard Kolić, and Mira Pavletić
and the Croatian Ministry of Culture for granting permis-
sion to study and publish the amphorae.
I am grateful for the support I have receive from the direc-
tors, past and present, of the Austrian Academy of Sciences,
Institute for the Study of Ancient Culture, Friedrich
Krin­zinger and Andreas Pülz.
I would like to take this opportunity to thank Piero Berni
Millet, Michel Bonifay, Claudio Capelli, Horacio González
Cesteros, Sándor Józsa, Alexander Schobert, Martino La
Torre and György Szakmány for their work. Thanks are also
offered to Claudio Capelli, Sándor Józsa and György Szak-
mány; they prepared the petrological (thin section) analyses
and the photomicrographs (1x1.3 mm) of the amphorae.
The petrological description of the amphorae can be found
in the catalogue. The technical assistance of Anna A. Nagy
and Ozren Grozdanić is very much appreciated.
BEZECZKY_Text.indd 12
I am especially grateful to Péter Hornung for creating
the FileMaker database. The amphorae, which are in an
extremely fragmented state, will be uploaded and saved in
the database.
Special thanks are due to Ágnes Bezeczky who translated
parts of the manuscript (Chapters 1, 2, 4 and 13).
The photos were taken by Andreas Pülz, Martino La
Torre and Tamás Bezeczky. For their permission to repro-
duce pictures which appear in the text we are grateful to Bri-
juni National Park.
The line drawings were pepared by Tamás Bezeczky, Piero
Berni Millet and Horacio González Cesteros. Thanks are also
due to Helka Németh who made the digital amphora draw-
ings for publication. The drawings and photos are actual size
on the plates. The scale of the rubbings is 1:1. Numbers in
bold refer to the catalogue in the chapters on the amphorae.
While I was working on the amphorae, the archaeologists
of many museums generously helped me: David Peacock (†),
Paul Arthur, Martin Auer, Marianna Bressan, Silvia Cipri-
ano, Ines Dörfler, Smiljan Gluščević, Jana Horvat, Simon
Keay, Ida Koncani Uhać, Vladimir Kovačić, Goranka Lipo-
vac Vrkljan, Stefania Mazzocchin, Péter Pánczél, Elena
Quiri, Federica Rinalani, Florian Schimmer, Eleni Schindler-
Kaudelka, Katalin Vanicsek, Paola Ventura, Péter Véninger,
Reinhold Wedenig and Susanne Zabehlicky-Scheffenegger.
Vienna, 2017 Tamás Bezeczky (†)
18.03.19 14:39
 XIII

LIST OF FIGURE S

Chapter 1 3.3 Monolithic platform of the early Christian altar.

ISTRIA AND THE AMPHORA WORKSHOPS 3.4 Triumphal arch supported by two monolithic col-
umns with capitals and imposts.
1.1 Map of Istria (Tabula Imperii Romani).
3.5 Capital in the sanctuary with crux coronata emblems.
1.2 Map of Fažana workshop and kiln.
3.6 Presbytery with septum.
1.3 Dolium from Brijuni harbour, the position of the
3.7 Hospitium (domus presbyterorum), northern annex.
dolia in the Verige and Castrum villas.
3.8 Room with double apses (cellae memoriae), southern
1.4 Complete Laecanius amphora from Magdalensberg.
annex.
1.5 The distribution of the Fažana amphorae.
3.9 Sarcophagus, narthex.
3.10 The Church of St. Peter, main entrance.
Chapter 2 3.11 Peutinger Table pars V.
THE VILLAS OF BRIJUNI
2.1 Map of Brijuni Islands and Fažana. Chapter 10
2.2 Map of the villa in Verige and the terrace building. CALENDAR GRAFFITI ON DRESSEL 20
2.3 Map of the terrace building. AMPHORAE
2.4 Map of Kolci villa.
10.1 A Dressel 20 globular amphora with a stamp, tituli
2.5 Castrum villa after Mlakar 1975-76 and
picti and graffito.
Begović – Schrunk 2007a.
10.2 Stages in the manufacture of the Dressel 20 amphora.
2.6 Castrum villa, after Mlakar 1975-76.
10.3 Nominal cursive graffiti.
2.7 Archive pictures from the excavations.
10.4 Correlations between stamps and binominal graffiti
2.8 The first Roman villa.
on pottery from Villar de Brenes.
2.9 Map of the first Roman villa.
10.5 Calendar graffito from the first group.
2.10 Map of the second Roman villa.
10.6 Three calendar graffiti from the second group.
2.11 The restored presses in the second villa.
10.7 Calendar graffito by Lucrio with a consular date of
2.12 In the lacus, spicae and opus signinum wall.
23 June 158 A.D.
2.13 The oil cellar and fragmented dolia in the second
10.8 Calendar graffiti from the fourth group.
villa.
10.9 An unpublished graffito from the fourth group.
2.14 Mill stone and a mola olearia.
10.10 The large globular body of the Dressel 20 amphora
2.15 Map of the third Roman villa.
from Brijuni with an ante cocturam calendar graffito
2.16 The smaller cellar and the dolia.
on its base.
2.17 Map of the Late Roman and Byzantine villa.
10.11 The disposable upper part of a Dressel 20 amphora
2.18 Wine presses installation.
found in Nijmegen and Augst and Kaiseraugst.
10.12 Calendar graffito on the base of a Dressel 20 ampho-
Chapter 3 ra from Brijuni.
THE EARLY CHRISTIAN CHURCHES OF 10.13 Brijuni calendar graffito.
BRIJUNI 10.14 Geographical distribution of the Asiaticus graffiti.
10.15 The Altenstadt graffito.
3.1 The Church of St. Mary.
10.16 The Lincoln graffito.
3.2 Column of the main nave with impost and Greek
10.17 The Wall graffito.
cross.

BEZECZKY_Text.indd 13 18.03.19 14:39

 XIV List of figures

10.18 The Brough on Noe graffito. Chapter 12

10.19 The Saint-Thibéry graffito. MICROPETROGRAPHY OF THE
10.20 The Reims graffito. FAŽANA AMPHORAE
10.21 The Béziers graffito, the Vindolanda graffito.
12.1 Geological map of Istria with field sampling
10.22 Melander graffiti: (1) Adelfa, (2) Testaccio.
locations.
10.23 Definite dates for all the months of the year.
12.2 Terra rossa covers the Mesozoic limestone in
10.24 Titulus δ and the stamp on a Dressel 20 amphora
the coastal zone of Istria.
manufactured during the time of Caracalla (212-217
12.3 Outcrop of flysch rock series from the interior
A.D.) and used in the year 224 A.D.
of Istria, north of Zajci.
12.4 Raw material transported from Trieste bay.
Chapter 11
INSCRIPTIONS ON LATERITE FINDS
FROM BRIJUNI
11.1 The final letters EVPOR visible in the retrograde
stamp on imbrex.
11.2 Post cocturam capacity graffiti on the dolia from Val
Catena.

BEZECZKY_Text.indd 14 18.03.19 14:39


LIST OF PL ATE S
THE AMPHORAE
Plate 1 Lamboglia 2 (nos. 1-3), Dressel 6A (nos. 4-8), ante
Dressel 6B (nos. 9-10) amphorae.
Plate 2 Dressel 6B (nos. 11-24) amphorae.
Plate 3 Dressel 6B (no. 25), Fažana 1 (no. 26), Fažana
2 (no. 27), Porto Recanati (nos. 28-29), Forlim-
popoli (nos. 30-31), Aquincum 78 (nos. 32-33)
amphorae.
Plate 4 Dressel 2-4 (nos. 34-36), Dressel 7-11 (no. 37), Bel-
tran IIA (no. 38), Dressel 14 (no. 39), Dressel 20
(nos. 40, 42-45), Almagro 50A/Key XXII (no. 46),
Almagro 51A-B (nos. 47-48) amphorae.
Plate 5 Dressel 20 (no. 41), Almagro 51C (no. 49), Beltran
68 (nos. 50-51) amphorae.
Plate 6 Gauloise 4 (no. 52), Mid Roman 1 (nos. 53-54),
Palatine East 1/LRA 1 (nos. 55-56), Crete TRC 2
(no. 57), Rhodian (nos. 58-60), Knidian (no. 61),
Koan with pinched handle (no. 62), Koan (nos. 63-
64) amphorae.
Plate 7 Koan (no. 65), Dressel 5 (no. 66), Crétoise AC4
(nos. 67-69), Agora G 199/Pinched handle (no.
70), Agora F 66 (no. 71), Agora M 125 (no. 72),
Agora M 279 similis (no. 73), Agora M 126 (nos.
74-76), Agora M 240 (no. 77), Kapitän I (no. 78)
amphorae.
Plate 8 Kapitän II (no. 79), Agora M 273 (no. 80), Late
Roman Amphora 1 (nos. 81-86), Late Roman
Amphora 2 (nos. 87-88), Late Roman Amphora 3
(nos. 89-90), Ephesus 56 (no. 91) amphorae.
Plate 9 Ephesus 56 (no. 92), Agora M 307 (nos. 93-96),
Late Roman Amphora / Gaza (nos. 97-99), Late
Roman Amphora 13 (no. 100), Samos Cistern (no.
101), Beirut (no. 102), Cretan AC1C / MRC 3
(no. 103), M 235/6 (no. 104) amphorae.
Plate 10 Late Roman Amphora 7 (no. 105), Sinope C (nos.
106-107), Hammamet 3 (no. 108), Tripolitanian
III? (no. 109), Africana I A/B (no. 110), Africana
I variant? (no. 111), Africana II A? (no. 112), Af-
ricana II B, Pseudo-Tripolitanian? (no. 113), Afri-
cana II C3 (no. 114), Africana II C (no. 115), Af-
BEZECZKY_Text.indd 15
XV
ricana II D (no. 116), Africana II/III transitional
(no. 117), Africana II variant (no. 118) amphorae.
Plate 11 Africana II variant (no. 119), Africana III A (nos.
120-122), Africana III A/B (no. 123), Africana
III B (nos. 124-126), Africana III C (no. 127),
Spatheion 1? or Keay 35A (no. 128), Spatheion
1 (nos. 129-130), Spatheion 1 late (no. 131) am-
phorae.
Plate 12 Spatheion 3A (no. 132), Spatheion 3B (no. 133),
Spatheion 3C (nos. 134-136), Keay 3/5 or 64
(no. 137), Keay 8B (no. 138), Keay 59 and Keay
8B (no. 139), Keay 11B var. Keay 1984, Fig. 172.2
(no. 140), Keay 35A (no. 141), Keay 35B (no. 142),
Keay 36 (no. 143), Keay 55 (nos. 144-145), Keay
57 (nos. 146, 147) amphorae.
Plate 13 Keay 57 (no. 148), Keay 57-56-55 (no. 149), Keay
62 Q or Albenga 11/12 (nos. 150-152), Keay 62
(nos. 153-154), Keay 62 or 61? (no. 155), Keay 61
(nos. 156-157), Keay 61A/D (nos. 158-159), Keay
61? (no. 160), Keay 61C (no. 161), Keay 8A (nos.
162-163) amphorae.
Plate 14 Keay 34 (nos. 164-166), Keay 1B ? (no. 167), Stor-
age jars? (nos. 168-169), Miscellaneous (nos. 170-
171, 173-178) amphorae.
Plate 15 Miscellaneous (nos. 172, 179-184) amphorae.
Plate 16 Lamboglia 2 (nos. 2-3), Dressel 6A (nos. 5, 7),
Dressel 6B (nos. 11-19) amphora stamps.
Plate 17 Dressel 6B (nos. 20-22), Beltran 2A (no. 38), Dres-
sel 20 (no. 40), Late Roman Amphora 1 (no. 86),
Late Roman Amphora 13 (no. 100), Hammamet
3A (no. 108), Africana III A/B (no. 123).
Plate 18 Photomicrographs Dressel 6B (nos. 9, 11-17).
Plate 19 Photomicrographs Dressel 6B (nos. 18-22), Bel-
tran 68 (nos. 50-51), Hammamet 3A (no. 108).
Plate 20 Photomicrographs Africana I A/B (no. 110), Af-
ricana I variant (no. 111), Africana II/III transi-
tional (no. 117), Spatheion I ? or Keay 35A (no.
128), Spatheion I late (no. 131), Spatheion 3A
(no. 132), Spatheion 3B (no. 133), Spatheion 3C
(no. 134).
18.03.19 14:39
 XVI List of plates
Plate 21 Photomicrographs Spatheion 3C (nos. 135-136),
Keay 3/5 or 64 (no. 137), Keay 11B var. Keay 1984,
Fig. 172.2 (no. 140), Keay 35A (no. 141), Keay 36
(no. 143), Keay 57 (nos. 146-147).
Plate 22 Photomicrographs Keay 62 Q or Albenga 11/12
(nos. 150-151), Keay 62 (no. 153), Keay 61? (no.
160), Keay 34 (nos. 164-166), Keay 1B? (no. 167).
Chapter 11
INSCRIPTIONS ON LATERITE FINDS FROM
BRIJUNI
Plate 23 [M·ALBI·R]VFI (no. 1), IMP·AVG·G[ER]
(no. 2), L·BARBI·L·L·EVPOR[IS] & L·BARBI·
L·L·EVPOR[I] (no. 3), Q·G·NICEP (nos. 4a1,
4a2, 4a3, 4a4, 4a5), C·CEIONI·MAXI (nos. 5a1,
5a2), Q·CLODI·AMBROSI (no. 6a1).
Plate 24 Q·CLODI·AMBROSI (no. 6a2, 6a3, 6a4, 6a5,
6a6), A·FAESONI·AF (nos. 7a1, 7a2, 7a3, 7a4).
Plate 25 A·FAESONI·AF (nos. 7b1, 7b2, 7b3, 7b4, 7b5,
7b6).
Plate 26 A·FAESONI·AF (nos. 7b7, 7b8, 7b9, 7b10, 7b11,
7b12, 7b13, 7b14, 7b15, 7b16).
Plate 27 A·FAESONI·AF (nos. 7b17, 7b18, 7b19, 7b20,
7b21, 7b22, 7b23-Gnirs 1908, 7b23-Matijašić
1987, 7b24, 7b25).
Plate 28 A·FAESONI·AF (nos. 7b26, 7b27, 7b28, 7b29,
7b30, 7b31, 7b32, 7b33, 7b34, 7b35, 7b36, 7b37).
Plate 29 A·FAESONI·AF (nos. 7b38, 7b39, 7b40),
C·FLAVI (nos. 8a1, 8b1, 8b2, 8b3, 8b4, 8c),
L·FVLLONI (nos. 9a1, 9a2-Gnirs 1908, 9a2-
Matijašić 1987, 9a3, 9a4, 9a5, 9a6, 9a7),
C·IVLI·AFRICANI (no. 10).
Plate 30 C·LAECANI·P·F (no. 11a ), LAEC (nos. 11b1,
11b2, 11b3-Gnirs 1904/1910, 11b3-Matijašić
1987, 11b4, 11b5, 11b6, 11b7, 11b8, 11c1, 11c2,
11c3, 11c4).
Plate 31 LAEC (no. 11d), PANSAE·VIBI (nos. 12a1, 12a2,
12a3, 12a4, 12a5, 12a6, 12a7).
Plate 32 PANSAE·VIBI (nos. 12a8, 12a9, 12a10, 12a11,
12a12), PANSIANA (12b, 12c1, 12c2, 12d1,
12d2).
Plate 33 TI·PANSIANA (nos. 12e, 12f ), C·CAESAR·
PANS (no. 12g), NER·CLAVD·P (no. 12h), PAN-
BEZECZKY_Text.indd 16
SIAN (no. 12i), L·PETR·AVIT (no. 13a), L·PETR
(nos. 13b1, 13b2), M·SERI (nos. 14a1, 14a2, 14a3,
14a4).
Plate 34 M·SERI (nos. 14a5, 14a6, 14a7, 14b1, 14b2, 14b3),
SISENNAE (no. 15), L·ST·IVSTI (no. 16), TRO-
SI (nos. 17a1, 17a2, 17a3), dolium stamp ARIA(…)
(no. 18), dolium graffito LXIII(…) (no. 19).
Chapter 12
MICROPETROGRAPHY OF THE FAŽANA
AMPHORAE
Plate 35 Polarizing microscopic photos 1. Fired terra rossa
soil (IST 12), 2. Chert grain in calcareous sandstone
(flysch, IST 64b), 3. Chalcedony sponge spicule in
calcareous sandstone (flysch, IST 64b), 4. Calcare-
ous sandstone (flysch, IST 64b), 5. Micritic clay-
stone (flysch, IST 31a), 6. Empty-shelled foramin-
ifera in recent marine sediment (IST 16d), 7. Boring
sponge traces on calcareous fragment in recent ma-
rine sediment (IST 17), 8. Opalic sponge spicule
(acicular) in recent marine sediment (IST 17).
Plate 36 Polarizing microscopic photos 1. Opalic sponge
spicule (sterraster) in recent marine sediment (IST
16), 2. Thin-shelled clam in recent marine sediment
(IST 67), 3. Fired recent marine mud (IST 68), 4.
Texture of typical amphora (Castrum 13), 5. Fired
terra rossa (IST 12), 6. Chert grain in amphora
(Castrum 14), 7. Microsparite filled globigerinida
foraminifera in amphora (Castrum 18), 8. Micrites
and mollusc skeleton fragment in amphora (Cas-
trum 12).
Plate 37 Polarizing microscopic photos 1. Opaque filled
globigerinida in amphora (Castrum 16), 2. Chal-
cedony sponge spicule and terra rossa inclusion
in amphora (Castrum 13), 3. Empty-shelled fora-
minifera and opalic sponge spicule in amphora
(Castrum 18), 4. Thin-shelled clam in amphora
(Castrum 14), 5. Opalic sponge spicule (acicular)
in amphora (Castrum 16), 6. Opalic sponge spicule
(sterraster) in amphora (Castrum 16), 7. Boring
sponge traces on calcareous fragment in amphora
(Castrum 12), 8. Terra rossa soil fragment in am-
phora (Castrum 22).
18.03.19 14:39
 SHORT HISTORY OF THE WORK SHOP
AND THE VILL AS

BEZECZKY_Text.indd 1 18.03.19 14:39

BEZECZKY_Text.indd 2 18.03.19 14:39
 3

1 ISTRIA AND THE A MPHOR A WORK SHOPS

Ta m á s B e z e c z k y

1 HISTORICAL BACKGROUND
Istria is a large peninsula located in the northeastern part of
the Adriatic region. The Roman Senate considered the pos-
sibility of a military occupation of Istria several times dur-
ing the second century B.C. This was one of the motives for
founding the town of Aquileia.1 The town, which was built
at a strategically advantageous location, became the most sig-
nificant trade centre at the head of the Adriatic. It also served
as an important military base for the northern campaigns. It
was in the interests of the Aquileian troops as well as the Ital-
ian merchants to develop a political and economic relation-
ship with the areas north of the Alps. They had an excellent
connection with the kingdom of Noricum, as evinced by the
fact that the king Voccio sent three hundred Norican caval-
rymen to support Caesar in the civil war against Pompey.2
The respectable Italian settlers who came in the wake of
Julius Caesar’s conquest promoted the Romanization of the
region; Tergeste and Pola were granted the rank of colonia.
Octavianus-Augustus’ interest in the region started after the
war against the Iapodes in 35-33 B.C. The advantages of the
peninsula were discovered during this period. The ties between
Italy and Istria developed quickly. As a result, the peninsula
became part of Regio X (Venetia et Histria) between 18 and
12 B.C.3 Some believe that Istrian oil production replaced the
production in Apulia during the late Republican period. The
Apulian region went through an economic crisis, the result of Fig. 1.1 Map of Istria (Tabula Imperii Romani).
which was a decline in oil production in Brindisi.4 The politi-
the villas cultivated, harvested, and processed the olives (the
cal changes offered economic opportunities for the new elite.
ancient sources regarded Istrian olive oil as one of the best
Some of the investors probably came from southern Italy.
on the market) and wines.7 They also had their own presses,
The traces of several villas have been discovered from Terg-
este to Pola.5 The owners of the villas were mainly senators
and members of the Roman elite.6 The Roman proprietors of 7
Pliny the Elder mentions the olive oil from Istria and Baetica in his
NH (15.9). Between 77 and 79 when he wrote the book, olive oil
was only produced in Istria; production in northern Italy had already
ceased. Thus, he praised only the oil of Venafrum. A little later, in
1
Livy 40, 34. about A.D. 102, Martial (12.63) mentioned the olive oil from Istria
2
Caesar BC 1, 18, 5. and Baetica in one of his epigrams. He also considered the Venafrum
3
Degrassi 1953, 54. oil to be the best. Pausanias (10.32.19) in his travel book compared
4
Baldacci 1967-68, 14; Manacorda 1995, 177. the Tithorea oil with the ones from Istria and Baetica. This descrip-
5
Matijaišić 1988, 27-71, T 2-3. tion from the second century may have relied on either Pliny or Mar-
6
Tassaux 1983-84, 193-229. tial (or some other source), because at that time only Baetican or

BEZECZKY_Text.indd 3 18.03.19 14:39

 4 SHORT HISTORY OF THE WORKSHOP AND THE VILLAS

Fig. 1.2 Map of Fažana workshop and kiln, I. Map of Fažana, after Gnirs 1910: A - Sv. Marija Church. B - The site where the amphorae were found.
C - The site of clay depots. D - Harbour. E - Amphora kiln. F - Ruins of the ancient villa rustica. G - Piazza del Duomo. H - The site where the
Laecanius amphorae were found. K - Church, 1-2-3 Excavation 1990-1991; 4 - Trench with three floor levels (Bezeczky 1998a, 4); II. Map of
the excavation 2007-2009: after Bulić – Koncani Uhač 2011: T1 - Plan of the amphora kiln 1, T2 - Plan of the amphora kiln 2.

BEZECZKY_Text.indd 4 18.03.19 14:39

 CHAPTER 1 – Istria and the amphora workshops (Bezeczky)
storage cellars, and ceramic workshops. However, there are
only two excavated amphora workshops in the region. One
figlina has been found near Poreč (Parentium) in the bay of
Loron,8 and the other is in the north of Pula in Fažana. Stone
inscriptions and roof tiles point to other workshops at Novi-
grad, near Piran, and near Trieste, which may have produced
amphorae.9
2 THE FAŽANA WORKSHOP
Anton Gnirs’ excavations at the beginning of the twentieth
century found a kiln site under the modern buildings, marked
it on a map and published the stamped amphorae.10 How-
ever, it seems probable that the description and the map were
made only in connection with a new water main (plumb-
ing). The excavations of 1990-1991 used similar methods.
The excavated floor layers provided entirely new information
regarding the possible existence of a villa to the south of the
workshop.11 Since this area is separated by a cistern from the
workshop and the kiln, it is possible that there was a villa to
house the workers of the workshop. This area may have been
isolated from the workshop. Gnirs mentioned the ruins of a
press and a dolium north of the kilns.12 They must have been
related to the production of olive oil. Another villa is sup-
posed to be here but there is no precise information. Our
knowledge increased considerably when the reconstruction
of Fažana’s road network began in 2007. The excavations
African oil was still available on the markets. Galen of Pergamon also
mentioned Istrian oil at the end of the second century (Buonopane
2009, 27-28). Finally, it is worth mentioning the letter of Cassiodorus
(Variae 12, 22-24) from 537/538. He writes about the plentiful har-
vest of oil, wine and wheat in Istria. They were shipped to Ravenna.
However, we do not know what kind of vessels were used. Nor is it
clear which part of Istria he was writing about.
8
A team consisting of Croatian, French and Italian members have been
working since 1994 in Loron. They unearthed the ruins of the work-
shop and four chambers of the kiln. No villas have been excavated
so far. The amphora stamps have the names of well-known persons.
The owners of Loron were T. Statilius T·F·Taurus Sisenna, later the
wife of Emperor Claudius MES·CAE (Mes(alinae) Cae(ari uxoris)
and Calvia Crispinilla. The emperors (Domitian, Nerva, Traian and
Hadrian); Marion – Starac 2001, 97-118; Tassaux 2001, 511-
512; Manacorda 2010, 217-227.
9
Tassaux 2001, 512-517, with detailed bibliography; TITAC (Tita-
tia Tertia), THAL (L. Quinctus Thallus), L·TERENTI (), C·ALTEN
(C. Altenus or Altenius), P·C·QVIR, TRAVL ·ET· CRIS, CRISPINI,
T·A·F· CRISPINAE; Koper: Bezeczky 1985, P·ITVR·SAB (P. Itu-
rius Sabinus); Cipriano 2009, 178.
10
Gnirs 1910a, 79-88; Gnirs 1910b, 102; Gnirs 1911b, 37-38.
11
The rescue excavation was in 1991, see Bezeczky – Pavletić 1996,
143-148, Fig. 3 and Bezeczky 1998a, 4, Fig. 3.
12
Gnirs 1910a, 81.
BEZECZKY_Text.indd 5
5
reports mention two kiln chambers, although most of the
ruins are below modern buildings.13
There are three periods of the Fažana workshop:14
- In the first period (from ca. 40 B.C. to A.D. 78), the
workshop was the property of the Laecanius gens. The Laeca-
nius family died out without a direct heir in 78 A.D.
- In the second period (from the Flavian period to the
period of Hadrian, from A.D. 78 to 138) both the property
and the workshop were taken over by the emperor Vespasian
and were integrated into res privata.
- During the third period (ca. the last third of the sec-
ond century to the early third century), the amphorae of
M. Aurelius Iustus were produced.
THE PRODUCTS OF THE WORKSHOP
The excavations make it clear that the workshop produced
dolia, amphorae, stoppers, tiles, spicae, heating pipes, and
clay lamps. The most important products of the workshop
were the amphorae. Only two of the kiln chambers are
known (Fig. 1.2, II). The material composition of the vessels
sheds light on the production technology. The petrological
analyses have shown that terra rossa, although available in
the vicinity, was not the basic inclusion.15 The bulk of the
raw material consisted of flysch. Only a small quantity of
terra rossa and recent marine sediment were mixed or added
as temper in Fažana. The geological study of the peninsula
shows that flysch is available in large quantities in the Koper/
Trieste bay. The Laecanii had huge properties near Trieste
(Materia). It seems probable that this area provided the
flysch necessary for the production. The amphorae filled
with olive oil were shipped to the north Adriatic ports and
Aquileia. Then the empty ships may have been filled with
flysch as ballast.16 The types of ships the Laecanii used are
not known, which makes it almost impossible to guess the
quantity and frequency of the shipments. It was during the
sailing season lasting from April to October that the neces-
sary quantities of flysch had to be collected. The organization
of the shipments must have been a major operation. Olive oil
was stored in dolia in the villas and only before shipping was
it filled into amphorae.
13
Bulić – Koncani Uhač 2009, 286-292; Bulić – Koncani
Uhač 2011, 123-128.
14
Bezeczky – Pavletić 1996, 147-148; Bezeczky 1998a, 4.
15
Józsa et al. 2016; see Szakmány – Józsa in this volume; previous
works: Mange – Bezeczky 2006; Mange – Bezeczky 2007;
Bezeczky – Mange 2009.
16
See Fig. 12.4, Szakmány – Józsa in this volume.
18.03.19 14:39
 6 SHORT HISTORY OF THE WORKSHOP AND THE VILLAS

3 THE OWNERS OF THE WORKSHOP: offered to Iupiter by C. Laecanius Ialysos (CIL, V. 14 = Inscr.
THE LAECANIUS GENS17 It. X. 630). There is another stone inscription in Guran, also
near Vodnjan (Dignano), which mentioned C. Laecanius
1. The first known member of the family in Istria was
Amycus and his matrimony Phorbe.22 The names Ialis (or
Publius Laecanius. He was born circa 90 B.C.
Iali) and Amycus occur on a number of amphora stamps
2. The son of Publius Caius Laecanius was also among the
and it seems reasonable that the persons mentioned in the
founders of Pula. His Dressel ante 6B amphorae and roof
inscription and the one who signed the amphorae are the
tiles have the C·LAECANI·P·F stamp.
same. The work of the workshop managers/vilici had to
3. His son Caius Laecanius was probably decurio. His roof
be organized. This may have been the job of C. Laecanius
tiles have the LAEC, his amphorae the C·LAEC·A stamp.
Menander, who was secretary to both the senators (CIL,
He was probably born between 55 and 35 B.C.
V. 8142 = Inscr. It. X. Reg. X, 1, 114).
4. His son Caius Laecanius Bassus was probably a knight.
His amphorae have the C·LAEC·BASSI stamp. His ampho-
rae are the first to have two stamps on the rim. (FELIX·SER). THE EMPERORS
The Magdalensberg layers date them to 10-5 B.C.
Paolo Baldacci mentioned that the figlina in Fažana was
5. Senator Caius Laecanius Bassus was praetor urbanus
taken over by the emperors after the family died out.23 The
in A.D. 32 and consul suffectus in A.D. 40. The fragment of
fact that the estate was taken over by the emperors can be
the sodales augustales claudiales mentions him as pater in the
established from the evidence the stamping system offers.
year 64.
The production of the amphorae with the emperor’s stamp
6. His son C. Laecanius Bassus was consul ordinarius in
was discontinued during Hadrian’s era.
64 A.D. The junior Laecanius died of anthrax in 78, which is
known from a remark of Pliny’s.18
7. The Laecanii adopted C. Laecanius Bassus Caecina M. AVRELIVS IVSTVS
Paetus. He was consul suffectus in 70 and proconsul of Asia
There is only one amphora with this name. M. Aurelius
Minor (in Ephesus) in 78.19 He had properties near Mintur-
Iustus may have been a freedman who was responsible for
nae, which are known from the inscriptions of his freedmen.
the workshop and the production in the Castrum villa.
8. His son C. Laecanius Bassus Caecina Flaccus died in
There is a votive altar dedicated by M. Aurelius Iustus to
Brindisi when he was 18.
the goddess Flora in the cemetery near the villa.24 We have
9. C. Laecanius Bassus Paccius Paelignus is also known.
no more information about this person. He may have been
He had a dedication to Laecanius (6).
a tenant (conductor) who paid rent to the Imperial treasury.
There is an important stone inscription from the period
of Claudius that documents the property of the senator 4 THE AMPHORA STAMPS
Laecanius pater (5). The inscription (CIL, V. 698 = ILS
We have to update our knowledge and prepare a distinct
5889 = Inscr. It. X. 4, 376) mentions a property in northern
chronology of the Laecanius and Imperial amphorae. Res-
Istria at Materija next to the Rundictes tribe (today part of
cue excavations took place in Fažana, in the course of which
Slovenia).20 This area used to be hills covered with woods,
new stamps were found and the kiln was excavated.25 Many
suitable only for animal farming and the timber industry.
amphorae with Laecanius and Imperial stamps have been
The Laecanius family must have had other properties. In
found outside Istria since 1998.26 The list of the ca. 1 500
Pula the Val San Pietro finds may be pieces of evidence.21
There is a stone inscription in Sv. Mihovil Bajolski (San
Michele di Bagnole) near Vodnjan (Dignano) which was 22
Matijašić 2001, 342-344.
23
Baldacci 1967-68, 34.
24
ILJug (Inscriptiones Latinae quae in Jugoslavia) 1204; Mlakar 1979,
17
I shall follow the way Tassaux 1998, 83 has reconstructed the Lae- 23-24; Starac 1995, 138, note 27.
canius family tree. 25
Paić – Bulić 2008, 17-40; Bulić 2009, 257-270.
18
Pliny NH 26.5. 26
Belotti 2004; Bezeczky 2005b, 49-50; Bezeczky 2014; Bulić
19
Taeuber 2011; Bezeczky 2013, 221. 2009, 264; Cipriano 2003; Cipriano 2008; Cipriano – Fer-
20
CIL, V. 698 = ILS 5889 = IIt X 4, 376; Slapšak 1977, 122-128; rarini 2001; Džin – Šalov 2008; Gabucci – Quiri 2008;
Tassaux 1982, 248; Milotić – Petrak 2012, 297-310. Gostenčnik 2002; Gugl 2003; Gugl 2004; Mazzocchin 2006;
21
Gnirs 1910b. Mazzocchin 2007; Mazzocchin 2010; Mazzocchin 2013;

BEZECZKY_Text.indd 6 18.03.19 14:39

 CHAPTER 1 – Istria and the amphora workshops (Bezeczky)
stamps can be updated almost on a daily basis. They provide
us with an increasingly precise understanding of the Fažana
workshop production.
In the late Republican period, an amphora rim with
a single stamp C.LAECANI·P·F belongs to the earliest
Istrian Dressel 6B form (ante 6B). The shape of the Laeca-
nius amphorae changes somewhat and the classic Dressel 6B
shape develops during the Augustan period. Every amphora
produced in the Laecanius workshop at that time had two
stamps on the rim. The stamp of Laecanius is at the centre
(C·LAEC·BASSI), with the second stamp (FELIX·SER)
above the handle. Later, from the Tiberian to the early Clau-
dian period, the C·LAEK·BASSI stamp uses the letter “k”
instead of a “c”. The Laecanii were the definitive producers
and transporters of Istrian olive oil for over one hundred
years.
From the Tiberian to the early Claudian period (Mag-
dalensberg II) there are 41 different stamp types. The
C·LAEK·BASSI stamp and its variants combined with the
workshop manager / vilicus names (e.g. A, ADEL, ARCI,
BAR, BARB, CAESI, CAR, CLARUS, COM, COMI,
EUCHARISTI, FA, FAV, FELIX, FVI, HER, HERME,
HOM, IALIS, IALI, L, OPI, OPTA, SPERATUS, SYNT,
VIAT). Post Magdalensberg period I we know of 15 stamp
types. The workshop managers’ / vilicus names are different
from the previous period (e.g. AMETHYSTI, AMYCUS,
BAR, CRESCENTIS, DI, EUCHARISTI, FAL, MARTI,
NICOMEDE, PIERI). Post Magdalensberg period II the
amphorae produced then can be dated with some precision.
Before the death of Laecanius, we know of 4 vilicus /
workshop manager stamps (CLYMEN, DATI, PAGANI,
PTOLEM). These stamps also co-occur with the stamp of the
emperor Vespasian. These are the pairs of stamps used dur-
ing the reign of Vespasian: IMP (CLYME, CLYMEN, DAT,
PAGANI, POLL), IMPE·VESP (BARNAE, PAGANI),
IMP·VESP (COLONI, PAGANI), IMP·CAES·VESP
(DAT, PTOLEM), IMP·CAES·VESP·AVG (CLYMEN).
Titus: IMP·T and IMP·T·CAES·AVG; (BERENTS),
IMP·T·CAES·AVG (PRIMIGEN); Domitian: IMP·
DOMITI and IMP·DOM (LESBI); only one stamp is
known from the time of Nerva: IMP·NERVE; Trajan:
IMP·TRA (the stamp cannot be read with certainty:
Paić – Bulić 2008; Pesavento Mattioli et al. 2000; Quiri
2009; Quiri 2011; Sakl-Oberthaler 2000; Schimmer 2009;
Schindler-Kaudelka 2000; Starac 2001; Starac 2010a;
Tiussi 2008; Tiussi 2010; Toniolo 2011; Vidošević 2003.
BEZECZKY_Text.indd 7
7
MANI?; SERV, and there is an incomplete stamp …VS);
Hadrian: IMP·HAD.27
The stamp of M. Aurelius Iustus is very different from
the stamps of the Laecanian and Imperial amphorae. This is
a stamp in hollow lettering, in the genitive case. Only one
similar stamp was found during the new Fažana excavation:
F♠A with a pattern of leaves.28
The list of the stamps is in the APPENDIX.
5 THE PRODUCTION CAPACITY OF THE
VILLAS AND THE DISTRIBUTION OF THE
FAŽANA AMPHORAE
As we shall see later, three of the villas with olive cellars have
been excavated on Brijuni. The dolia in the cellars may be
suitable for finding out the volume of the olive oil produc-
tion.
There are a number of fragmented dolia in the cellars.
The capacity calculations were based on a complete dolia
exhibited on the island (Fig. 1.3). Its form and material cor-
responds to the pieces in the cellae olearia in the Verige and
Castrum villas. The capacity is approximately 1750 litres.29
Gnirs found numbers on the side of some of the dolia.30
The numbers may have referred to the quantity of the filled
amphorae. However, we do not know when or who wrote
these numbers on the dolia.
The villas thus had the following capacities: Verige / Val
Catena 56, Kolci / Monte Collisi 100 and Castrum 48 dolia.
Accordingly, the island produced about 357 000 litres of
olive oil per annum, plus the unknown quantities of Fažana
and Dragonera. This is, of course, a rough estimate. Around
9 400 amphorae would be needed to transport 357 000 litres
of olive oil. It should also be added that the production may
have fluctuated from year to year.
The capacity of the Dressel 6B amphorae was 37.9-39
litres. However, only a few complete amphorae are known.31
27
Published from a private collection by Starac 1994-95, 162, T 12.1.
28
Paić – Bulić 2008, 28, no. 34.
29
I am grateful to Martino La Torre for the information: the ceramic
capacity is approximately 403 l, which, if the gross density is 1.6,
means that its weight when empty was around 645 kg. Thus, the
dolia stood on a gravel bed to dry, and the convex shape of the floor
was preserved.
30
Gnirs 1908b, 179; Bezeczky 1998a, 71-72; see also Berni Millet in
this volume.
31
Ca. 12 congius = 39.3 litre. Complete amphorae Novaria: Bezeczky
1998a, nos. 3, 195, 197, 361 and 405; Carreum Potentia: Riva 1987,
92, Fig. 26, Pl. Va. 4 and Fig. 25, Pl. Va, 6a-b = Bezeczky 1998a,
nos. 439 and 624; Opitergium: Cipriano – Ferrarini 2001, 49,
18.03.19 14:39
 8 SHORT HISTORY OF THE WORKSHOP AND THE VILLAS

Fig. 1.3 Dolium from Brijuni harbour, the position of the dolia in the Verige and Castrum villas
(Drawing: La Torre 2017, Photo: Bezeczky 2014).

BEZECZKY_Text.indd 8 18.03.19 14:39

 CHAPTER 1 – Istria and the amphora workshops (Bezeczky) 9

Fig. 1.4 Complete Laecanius amphora from Magdalensberg; stamps: C·LAEK·B /

FELIX·PET (Drawing: La Torre 2017, Photo: Bezeczky 2014).

The height of the amphorae was 87-92 cm (Fig. 1.4). The
outer diameter of the 604 amphorae studied was 12-17 cm.
The most common diameter of the rims was 14-16 cm. This
is true of 85 % of the vessels. This also means that the potters
followed the model quite closely.
We have little information about the size of the Fažana 1
and Fažana 2 amphorae. Only the upper parts are still
around. The rim of the Fažana 1 amphorae was 12-14 cm,
while that of the smaller Fažana 2 amphorae was 8-9 cm. We
don’t have complete amphorae in Fažana and Brijuni, which
makes their capacity unknown.
The volume of production can only be guessed at on
the basis of the amphorae which were found in more than
75 sites over the last 100 years. The Dressel 6B amphorae can
be found in northern Italy along the Po valley all the way to
Torino.32 In Raetia, the transportation route of the ampho-
rae from Fažana can be traced to Curia, Bregens, Oberstim.
In Noricum and Pannonia, along the Danube, Istrian oil
was used all the way up to Aquincum in the north and Rit-
tium in the south. In more central areas, the most important
places where the amphorae have been found Magdalensberg,
Aguntum, area Siscia, Sirmium and settlements along the
Amber Route.33

nos. 51, 54, 57, 58, 62, 65, 78 and 91; Patavium: Pesavento Mat-
tioli et al. 2000, 37-38, Figs. 4, 7; Cipriano – Mazzocchin 32
Gabucci – Quiri 2008, 68-69; Mazzochin 2013, 78-81; Cipri-
1998, 365, no. 5, T 1, 5; Ateste: Cipriano – Mazzocchin 1998, ano 2009, 173-185; Bezeczky 1998a, 75; Bezeczky 2014, 247-
365, no. 2, T 1, 2; Vicentia: Mazzocchin 2013, 133, 25.11e; Mag- 252.
dalensberg: Schindler-Kaudelka 2000, 390, G1. 33
Bezeczky 1998a, 75, Fig. 47.

BEZECZKY_Text.indd 9 18.03.19 14:39

 10 SHORT HISTORY OF THE WORKSHOP AND THE VILLAS

Fig. 1.5 The distribution of the Fažana amphorae (Drawing: Bezeczky 2015).

BEZECZKY_Text.indd 10 18.03.19 14:39


2 THE VILL AS OF BRIJUNI
M artino L a Torre – Tamás Bezeczky
Pliny the Elder called the Brijuni Islands the Pullariae, a
name also used by the Peuntiger Table.1 The Brijuni Islands
are a group of fourteen islands. The largest island is Veli Bri-
jun (579 hectares).2 Eight villas on the island of Brijuni and
the surrounding islands were identified. Unfortunately, it
was not possible to excavate more than three villas (Verige –
Val Catena, Kolci – Monte Collisi and Castrum). Another
villa on the mainland, 5 km north of Fažana, called Drago-
nera, was recently explored. The amphora stamps suggest
that the villas formed part of the Laecanius family’s proper-
ties.3 Another villa that was excavated in 1909-1912 by Gnirs
is situated ca. 700 m south of Fažana Center surrounding a
60 m-deep bay called Val Bandon almost directly opposite
the large complex in Val Catena on the Brijuni island.4
1 VERIGE BAY (VAL CATENA)
The Verige villa (Val Catena), located in a bay, was a luxuri-
ous maritime terrace villa (villa maritima). Its length (from
east to west) measured approximately 450 meters. It may
have been possible to close off the bay with a chain (catena)
between the north and south moles, like in the Lion Harbor
of Miletus in Asia Minor.5 The complex was composed of
separate buildings standing in a row along the shoreline of
the narrow bay, which is more than 800 meters long. We find
the following buildings on the northern shore (from east to
1
Pliny NH 3, 26, 151; Peutinger Tablet segm. IV.1; Geographus
Ravennas (V, 24, 101 = no. 165).
2
Veli Brijun or Brioni Grande; in the text we used the current designa-
tion Brijuni.
3
A rescue excavation took place at the Dragonera villa located 5 km
north of Fažana, near Peroj, Starac 2010b. This building is a typi-
cal villa maritima, located on the seashore. It has its own port as well
as agricultural and residential sections. The southern wing of the
villa, located on the shore, was intended for residential purposes. The
western and eastern wings of the complex both contain facilities for
pressing olives. There are amphora rims with Laecanius and Imperial
stamps.
4
Gnirs 1911d, 155-186.
5
It is also conceivable that the term dates back to the Roman period,
as the chain would presumably have long been lost in the period after
Antiquity, Weisshäupl 1900, 204.
BEZECZKY_Text.indd 11
11
west): the north mole; an industrial area (fullonica?) with
water reservoirs and a small bath; the large thermae (F), with
the palaestra (G) and a vivarium (E) beneath the water. Fur-
ther to the west, we find a diaeta, an atrium (H) and a long
portico (K), with a cryptoportico behind it.
This made it possible to access the facilities on the north-
ern shore even when the weather was bad. Another diaeta
(L) leads to the temple area (M-P-N) in the western part
of the bay.6 The living quarters and the villa rustica on the
southern shore were connected to it by a peribolos hall in the
shape of a semicircle, which was behind the three temples.
This hall completed the temple area architecturally. There
was another cryptoportico beneath the hall.7 Gnirs found a
coin from the year 41 A.D. in the stonework of the hall.8
The main building (domus), in the southern part of the
bay, is what is called a terrace building (R).9 The building
had two parts. The western wing housed the living quarters,
while the eastern wing held the industrial area.10 A luxuri-
ous villa maritima was combined with a villa rustica here.
Unlike a villa urbana, which generally did not look imposing
from the outside and was organized around its courtyard,
these villas were open to the surrounding countryside. The
two peristylia were U-shaped and opened to the sea.
The industrial area included the oil presses in room X, but
only the channels were observable at the time of the exca-
vation.11 These channels connected the presses and the vat
6
Gnirs 1904, 145-146. There were more than a hundred Laecanius
amphorae in the debris of room L. The amphorae had the following
stamps: BARB, COM, COMI, IALIS and VIAT. All of them belong
to the first period of the workshop. The fragments of late Roman
amphorae (types 1 and 2) were found near the staircase of room F,
Bezeczky 1998a, 55.
7
Gnirs 1908b, 169-170.
8
Gnirs 1904, 140.
9
Gnirs 1915, 100-129.
10
Schrunk – Begović 2000, 255-257, assume that the industrial area
was built first, and the other parts of the villa were constructed in two
later phases.
11
Gnirs 1906, 33; Gnirs 1908a, 137 mentions a wine press and wine
cellar. He presented no arguments. By contrast, Rostovtzeff 1957,
237 speaks of oil production. Later Gnirs 1915, 119 mentions an
oil cellar (cella olearia); Schrunk – Begović 2000, 256, mention a
18.03.19 14:39
 12 SHORT HISTORY OF THE WORKSHOP AND THE VILLAS

Fig. 2.1 Map of Brijuni Islands and Fažana (Drawing: La Torre 2017).

BEZECZKY_Text.indd 12 18.03.19 14:39

 CHAPTER 2 – The villas of Brijuni (La Torre – Bezeczky) 13

Fig. 2.2 Map of the villa in Verige and the terrace building (Drawing: La Torre 2017).

BEZECZKY_Text.indd 13 18.03.19 14:39

 14 SHORT HISTORY OF THE WORKSHOP AND THE VILLAS
Fig. 2.3 Map of the terrace building (Drawing: La Torre 2017).
(lacus, 7.25 × 1.25 m, max volume 4.44 m3). The lacus was at
the end of the adjacent cellar. The walls of the vat were made
of opus signinum. It had an opus spicatum floor. There were
four rows of sunken dolia in the cellar. The columns which
held the roof divided the cellar into two naves. The stylo-
bate (the top edge of which has not been preserved) stood
30-50 cm above the threshold to the portico. The size of the
cellar was 8 × 27 m. It contained 56 (perhaps 60) dolia. The
narrow spaces on the bottom terrace of the villa rustica may
have served as a place to store amphorae.
There are two large water reservoirs (V) made of opus cae-
meticium on the top of the hill above the terrace-building. It
used the water from the karst well of Gradina-bunar (Z).12
The well was across from the terrace building. A small nym-
phaeum was erected next to the well. The water was lifted
with a bucket-chain system from the well and emptied into
two large reservoirs. These reservoirs were 18-20 m above
wine press and wine cellar. If the villa was the property of the Laeca-
nius family, it was undoubtedly used for olive oil production. There
are no amphorae for the production and export of wine. The same
is true of the other Laecanius villas. Only Dressel 6B amphorae pro-
duced locally have been found.
12
Bezeczky 1998a, 56, at the well of Gradina-bunar there were Dres-
sel 2-4, Lamboglia 2/Dressel 6A, Gauloise 4, Dressel 6B (with Laeca-
nius stamps), Forlimpopoli, Keay XXV, Spatheion, Almagro 50 and
LR 1 amphorae.
BEZECZKY_Text.indd 14
sea level, which provided sufficient pressure for the water
to reach the reservoirs on the hill on the southern side of
the bay above the main house. They were connected by lead
pressure pipes, which were installed in a canal.13 The two res-
ervoirs were constructed in two phases, from east to west;
the buttresses of the western reservoir were added in a third
phase. The eastern reservoir was divided into three sections.
This was necessary because major cracks in its side walls had
caused leaks. The new divisions were placed exactly where
the cracks were and sealed them. Both reservoirs were emp-
tied through an opening to the north. In addition to the
reservoir directly on the karst spring on the opposite slope
(Gradina-bunar), there were further water storage facilities:
two reservoirs to the northeast of the buildings, above the
north mole, which also had a karst well in front of them;
another reservoir to the northwest of the diaeta; and the cis-
tern underneath the peristylium of the villa’s industrial area.
In addition to the agricultural/industrial infrastructure,
there were three bathing facilities: a small bathhouse in the
northeast, at what is now the shoreline,14 the large thermae
across from the terrace villa and the small thermae to the
west of the living areas.
13
Weisshäupl 1899, 81-82; Gnirs 1908b, 174-175.
14
Gnirs 1915, 134.
18.03.19 14:39
 CHAPTER 2 – The villas of Brijuni (La Torre – Bezeczky) 15

Fig. 2.4 Map of Kolci villa, after Gnirs

1908 (Drawing: La Torre 2017);
A - living area, B - courtyard, C - oil
cellar, D - kitchen, E - oil presses,
F - lacus.

Thanks to the two springs, both the villa’s agricultural sec- 2 KOLCI (MONTE COLLISI)
tion and its luxurious facilities, including the wells and ther-
Gnirs found a villa rustica on Kolci hill (Monte Collisi),
mae, were as independent as possible from rainwater.
on the northwestern part of the island.20 The three wings of
An enclosed garden and the south mole were to the east
the villa were built around a central courtyard (B). The liv-
of the villa rustica.
ing quarters (A) were in the southwestern wing. The kitchen
The buildings on the northeastern part of the bay (A-D)
(D), the four oil presses (E), a vat (F) and the large cellar
were regarded both as the industrial area of the villa15 and as
(C) were to the east. Three adjoining rooms were arranged to
an independent villa rustica.16 The function of these build-
the southwest of the presses. Every press had a large (2.18 ×
ings is controversial. During the first period, the villa was
0.86 m) stone block.21 There were two rectangular holes in
fairly simple with industrial areas and two reservoirs.17 The
the stones for the wooden uprights (arbores). The diameter
glass and ceramic objects found here are fairly varied with
of the concave press beds (ara) was 1.9 m and it was made
respect to their chronology and type.18 In 1996-1997 Croa-
of small ceramic bricks (spicae). There was a stone channel
tian and German archaeologists were involved in underwa-
ter research in Verige bay, next to the Austrian pier.19
ber of amporae are misidentified, which makes his conclusions doubt-
ful. For example, some of the Fažana amphorae have not been identi-
15
Begović Dvoržak 1990, Fig. 1, no. 9. fied properly: nos. 950, 952, 967 are not Dressel 6A, they are rather
16
Gnirs 1915, 146-147; Matijašić 1982, 56-57. Fažana 1; no. 991 is Fažana 2; no. 960 is not ante Dressel 6B; nos.
17
Gnirs 1906, 29-32, 35-36; Gnirs 1915, 146-147. 979, 980 are not Dressel 6B; no. 1045 is Dressel 6B; no. 1090 is not
18
Bezeczky 1998a, 57, Lamboglia 2, Dressel 6A, Dressel 6B, Knid- Schörgendorfer 558; no. 1000 is perhaps Kapitän 2 and not Dressel
ian, Dressel 25 and Gauloise 4 amphorae were identified; Vitasović 9 similis; no. 915 is not Agora M 315, it is rather Agora m 279; no.
2004, 93-98 described some tile stamps (M·SERI, [A·FAE]SONI[·] 1095 is not Knossos 36 but rather Ephesos 56.
A[F] and L·PETRO), see Berni Millet in this volume. 20
Gnirs 1908a, 134-136.
19
Under the direction of Mario Jurišić and Helmut Bender; see Bloier 21
Gnirs 1908a, 135, mentioned four presses, of which only three have
2012. Unfortunately Bloier’s work is little help. In his book, a num- survived.

BEZECZKY_Text.indd 15 18.03.19 14:39

 16 SHORT HISTORY OF THE WORKSHOP AND THE VILLAS

Fig. 2.5 Castrum villa after Mlakar 1975-76 and Begović – Schrunk 2007a (Drawing: La Torre 2017).
1 - Prehistoric settlement, 2 - Remains of the salt works, 3 - Late Antique masonry tombs, 4 - Remains of the sarcophagus in situ,
5 - Church of St. Mary, 6 - Tombs in the narthex of the church, 7 - Late Roman settlement, 8 - Remains of the structures alongside
the salt works and quarries, 9 - Late Antique building, 10 - Location of the ara dedicated to the Goddess Flora, 11 - Quarry.

BEZECZKY_Text.indd 16 18.03.19 14:39

 CHAPTER 2 – The villas of Brijuni (La Torre – Bezeczky)
around it. This channel (12 × 4 cm) led the oil into the vat
(lacus, 8.65 × 2.30 m, max. volume ca. 13 m3) with an opus
spicatum base. The rectangular oil cellar in the eastern part
of the villa was evenly divided into two naves and contained
about 156 dolia, Relying on Cato’s estimates, there may have
been 100 olive oil dolia.22 The water supply of these facilities
is unknown.
Some objects were published from the villa: a vessel with
Barbotine decoration, one terra sigillata without any precise
definition, many roof tiles and stamped Laecanius Dressel
6B amphora fragments.23 There were Spatheion and Africana
III amphora fragments in the rubble of Gnirs’ excavation.24
3 CASTRUM
The Castrum villa is located on the western coast of Brijuni,
in a Madona bay (next to the small churches of St. Mary and
St. Peter). Saline Bay is located to the southeast of the villa.
A number of important articles were published in connec-
tion with the villa. The excavations in the Castrum villa were
started by Gnirs, continued by Mario Mirabella Roberti,
Štefan Mlakar and Anton Vitasović.25 Gnirs excavated St.
Mary’s Church, near the Castrum villa.26 As the villa excava-
tions continued, with breaks both short and long, for nearly
80 years, we do not have access to all of the data.
After Gnirs, Mlakar wrote the first longer article, which
appeared in Histria Archaeologica in 1976. He published a
detailed map, which was revised by Vitasović following an
excavation in the 1980s.27
The original documentation of the excavations was
unavailable, so we could only use the information provided
in the publications. However, we were able to study five
thousand photographs of the excavations in the Archive
of the Brijuni Museum. The layers were 1.6-1.8 m thick in
many places, which means that they were not suitable for
stratigraphic analysis.
22
Cato De Agr. 12, 10, mentioned a 240 iugera estate which produced
oil needing 100 dolia for oil, 10 dolia for amurca, 10 dolia for wine,
10 dolia for grape pomace, 20 dolia for corn.
23
Gnirs 1908a, 136; Gnirs 1908b, 177; C·CEIONI·MAXI and
[…?PAN]SIAN[A], see Berni Millet in this volume; Bezeczky
1998a, 62, Cat. 59, 154 and 155, C·LAEK·B, C·LAEK·B / [CO]M,
C·LAEK.
24
Found in the depot, Bezeczky 1998a, 62, inv. no. B 221.
25
Gnirs 1911a, 75-97; Mirabella Roberti 1935, 294-295; Mla-
kar 1975-76, 5-50; Vitasović 2005, 157-210; Vidrih Perko –
Pavletić 2000, 263.
26
Gnirs 1911a, 75-93.
27
Mlakar 1975-76, 50; Vitasović 2005, 181.
BEZECZKY_Text.indd 17
17
The material contained in various plastic bags was mixed
up in the course of storage and transportation. The fine
ceramics and glassware were stored separately, and we were
unable to examine them. An amphora bearing the stamp
FELIX·PET is missing from the finds as well as the tegula
with the stamp LAECANI·P·F.28 Due to the site’s long exca-
vation process, the missing documentation of the process
and the restoration work, which was carried out in conjunc-
tion with the excavation, the findings at the site can only be
interpreted in a very limited manner. A review of the docu-
mentation and the published articles shows that we do not
know enough about the history of the villa. Many questions
remain unanswered.
THE FIRST ROMAN VILLA
The earliest villa was found near the shore. Many researchers
dated this villa to the second century and the first century
B.C. They thought that this villa was probably destroyed in
the middle of the century. According to one opinion – based
on Mlakar’s excavations – there was a military camp on the
site of the villa during the Istrian wars of the second century
B.C. “Before the conquest of the Histrian hill forts Faveria
and Mutila on the Istrian mainland, and the destruction of
Nesactium in 177 B.C., the Roman army came from Aquileia
escorted by navy, which carried supplies. Madona bay, oriented
towards the west and shielded from the view from the main-
land, was the most logical location for the first Roman camp
and naval base on Brijuni. In the first century B.C., during the
civil war between Caesar and Pompey, such a military camp in
Madona Bay may have had a significant function. It is possible
that it suffered damage as a naval base in the course of battles
between Caesar’s and Pompey’s navies in which the numerous
Histrian population held the side of Pompey.”29
Based on the amphorae excavated at the Castrum site, it
is difficult to identify the earliest villa as a military camp. It
is not possible to determine the function of the rooms on
the basis of the walls of the villa. The published map does
not resemble a camp. Furthermore there are no amphora
finds from the second century (e.g. Greco-Italic, Dressel 1,
Brindisi amphorae). In a later article, the authors mentioned
above described the villa in an entirely different way:30 “The
earliest architectural finds in Dobrika Bay belong to a mod-
est rustic villa, dated by Mlakar to around 129-79 BC. Its
28
Inv. no. BA1867 and BA 4132; Matijašić 1998, 253.
29
Begović – Schrunk 2007a, 96.
30
Begović – Schrunk 2010, 252.
18.03.19 14:39
 18

BEZECZKY_Text.indd 18
SHORT HISTORY OF THE WORKSHOP AND THE VILLAS

Fig. 2.6 Castrum villa, after Mlakar 1975-76, A-Area of the big Roman villa, B-Area of the oblong building with a circular corner and a rectangular risalit at the side,
C-Area of the Late Antique and Byzantine buildings, D-Area of buildings from the Early Middle Ages and recently erected parts.

18.03.19 14:39
 CHAPTER 2 – The villas of Brijuni (La Torre – Bezeczky) 19

Fig. 2.7 Archive pictures from the excavations (Photo: National Park Archive).

Fig. 2.8 The first Roman villa (Photo: Bezeczky 2014).

BEZECZKY_Text.indd 19 18.03.19 14:39

 20 SHORT HISTORY OF THE WORKSHOP AND THE VILLAS
Fig. 2.9 Map of the first Roman villa (Drawing: La Torre 2017).
remains were found along the coast, below the level of a larger
rustic villa which used the two walls to construct its founda-
tions, leaving rooms of the earlier villa in its substructure. The
villa from the second century BC has a large production area,
which comprises approximately one third of its entire space, tes-
tifying to its production function.”
Vitasović formulated another very similar opinion:31
“The material remains and social circumstances date the villa
back to the period between the second half of the second century
BC and the first quarter of the first century BC, that is, to the
period between 129 and 79 BC. Even though the facility has
not been fully researched, the material remains and finds offer
an insight not only into dating but also into the owners’ various
activities and needs. The life necessities of the people who occu-
pied the villa were determined by the economic resources ( fer-
tile land, animals and the sea) which dictated the development
of agriculture, livestock breeding and fishery. This antique
enclave, which is relatively distant from any hill forts, existed
31
Vitasović 2005, 208.
BEZECZKY_Text.indd 20
until the time of the civil war between Anthony and Octavian,
when it was destroyed. In the first half of the first century AD, a
big Roman villa, 51 × 58.7 meters in size, was built above, to
the east and to the north of the above-mentioned villa.”
The period of civil war between Caesar and Pompey can
be connected more readily to the villa.
However, as we do not have any information regarding
the layers we do not know where the large quantity of Lam-
boglia 2 amphorae were unearthed. Regarding the coins, only
a single AS from the Republican period has been found.32
The published terra sigillata refer only to the Augustan era.33
We do not know who used the first villa in the beginning and
when it was abandoned.
32
Miškec 2002, 177; Vidrih Perko – Pavletić 2000, 265.
33
Vidrih Perko – Pavletić 2000, 263.
18.03.19 14:39
 CHAPTER 2 – The villas of Brijuni (La Torre – Bezeczky) 21

Fig. 2.10 Map of the second Roman villa (Drawing: La Torre 2017).

THE SECOND ROMAN VILLA villa. It is possible that some of their functions were fulfilled
jointly with the other Laecanius villas (Verige and Kolci).
The second villa was built above and behind the walls of the
Three presses were reconstructed at the highest point of the
first building. It was a U-shaped villa, open to the sea. Its
villa, next to a lacus and a cellar. We do not know what the
large courtyard was surrounded by a living area and the agri-
original form of the presses was; we only have a reconstruction
cultural buildings. The various representations of the living
made of stones. The measurements of the thus reconstructed
area are not identical.34 There is a small cistern in the large
presses match those excavated at Kolci (Monte Collisi) exactly.
central courtyard for collecting rainwater. There was clearly a
The bottom of the lacus is covered with spicatum (12 ×
well nearby, providing the inhabitants of the villa with water
6.5 × 2.1 cm). Its bottom band walls are made of 21 cm (in
to drink. As described by M. Porcius Cato, many things were
some places 23 cm) thick opus signinum walls, and this layer
needed in a villa rustica which produced olive oil: presses
insulates it from the stone walls. To the mortar of this layer
and dolia, agricultural equipment (cart, iron plough, har-
of opus signinum pozzuolana (Vesuvian area) and ceramic
row) and stables.35
fragments had been added to ensure the mineral hydration
There was not a great deal of space available at the villa to
to give a waterproof cover. The waterproof layer was needed,
house the work animals (oxen, donkeys, occasionally sheep)
because it prevented the oil and amurca from leaking out of
used for transportation and everyday work. The stables and
the lacus. This is exactly how the huge cisterns in the Cas-
other such buildings were presumably outside the walls of the
trum and Verige (Val Catena) villas and the basins near the
Gradina-bunar well were insulated.
There is a large cellar next to the lacus in the Castrum
34
Begović – Schrunk 2011A, 378; Vitasović 2005, 163-166.
villa. It is 8 m wide and 23.3 m long on the inside. The walls
35
Cato De Agr. 5, 2 and 12, 10.

BEZECZKY_Text.indd 21 18.03.19 14:39

 22 SHORT HISTORY OF THE WORKSHOP AND THE VILLAS
are 45 cm thick. The cellar is divided in the middle by a row
of columns. The dolia were sunk into the floor in four rows,
12 in a row, with 48 pieces in total. One dolium bears the
stamp ARIAL(...).36
Many modern authors who have written about the his-
tory of the villa assume, without any proof, that there was
wine production in the first century A.D.37 Locally made
vessels for transporting wine have not yet been found. How-
ever, the large quantity of Dressel 6B amphorae produced
in the Fažana workshop and found in the villa are the most
important evidence of oil production and transportation.
These amphorae bear the stamps of the owners (Laecanius
family and the Emperors). The situation is similar in the
other two villas in Brijuni. All of the buildings, presses, cel-
lars, dolia, and cisterns were made of the same materials and
based on the same concept.38
36
See Berni Millet in this volume.
37
Begović – Schrunk 2007a, 42 and 97 = Begović – Schrunk
2010, 255-256 = Begović – Schrunk 2011a, 377; Vitasović
2005, 209.
38
According to the micropetrographic analysis by Józsa and Szakmány,
the dolia of the Castrum villa show a variable composition and tex-
ture. The groundmass with fine-grained non-plastics is more or less
similar in both types, but the quantity of coarse temper is very dif-
ferent (approximately from 0 %-15 %). The groundmass of both dolia
types consists mainly of clay with a small amount of fine-grained
quartz, muscovite, strongly fired globigerinida and carbonate grains.
In most of the dolia, there is a considerable amount of non-plastic
material which is very large in size and consists of angular siliciclas-
tic material (dominantly quartz-quartzite, rhyolite, chert-radiolarite-
BEZECZKY_Text.indd 22
Fig. 2.11 The restored presses in
the second villa (Photo:
Bezeczky 2014).
The tegulae help to date the villa; they include various
stamps from the Pansiana workshop (PANSAE·VIBI) and
the stamps Q·CEIONI·MAXI and M·SERI. They were all
produced in the middle of the first century B.C., similarly
to the C·LAECANI·P·F tegula found here.39 One may
thus assume that the Laecanius family already possessed
the villa at the time. This is supported by a stamped Dres-
sel 6B amphora (earliest form). The stamp on the amphora
matches the tegula stamp perfectly (C·LAECANI·P·F). Fur-
ther tegulae with the LAEC stamp can be dated to the early
Augustan period, which indicates that the villa continued to
be in use.40 Only a few of the coins are dated to the Augustan
period.41
There are several mill stones inside and outside the cellar.
Some of them show oval wear. They belonged to the “stone
roller/broyeur á rouleau” type of mill. There are fragments of
spiculite, amphibolite, granite, flysch originated [partly slagish] sand-
stone clasts) and well rounded calcareous rock and terra rossa grains.
The groundmass of dolia is very similar (except for the lack of recent
marine opalic spiculae) to the material of the main type of Laeca-
nian amphorae, so their raw material must have a very similar source
(northwest Istrian flysch). The extremely large non-plastic material
was most probably artificially added as temper. This kind of material
with such a composition and grain size is not known in Istria.
39
See Berni Millet in this volume.
40
Perhaps connected with the stamped amphora C·LAECA, Pais 1884,
1077.86.
41
Miškec 2002, 177, one Dp 17 B.C. (RIC 342), two As 15 B.C. (RIC
382) and 7 B.C. (RIC 431).
18.03.19 14:39
 CHAPTER 2 – The villas of Brijuni (La Torre – Bezeczky) 23

Fig. 2.12 In the lacus, spicae

and opus signinum
wall (Photo: Bezeczky
2014).

a mill (mola olearia) near them.42 These were used during oil
production.
The earliest finds from the villa are the Lamboglia 2
amphorae, which contained Adriatic wine, and the Dressel
6B (ante 6B) amphorae, which contained olive oil. It is not
possible to determine to which phase of the villa they belong.
Many Dressel 6A, Dressel 6B and Dressel 2-4 amphorae, as
well as amphorae from Rhodes, Kos, Knidos and Crete were
found.43 These can be dated to the period from the late first
century B.C. to the second century A.D. Some coins are
dated from Tiberian to the Hadrianic period.44
Few stamped Dressel 6B amphorae from the Tiberian–
Claudian period have been found (11-17). We know of
only a few amphorae with an Imperial stamp in the period
from Vespasian to Trajan (18-22). However, these stamped
amphorae prove that the villa was continuously in use. This
is also supported by a number of unstamped Dressel 6B
amphorae. The crisis during Hadrian’s reign is indicated by
a Dressel 20 type stamped amphora (40) from Baetica. Why

42
Mlakar 1975-76; Bezeczky 1998a, 65; Vitasović 2005, 191;
Matijašić 2008, 292-293.
43
Vidrih Perko – Pavletić 2000, 265-266; see Bezeczky (Chapter
4) and González Cesteros in this volume.
44
Miškec 2002, 177-178, Tiberius two As 15-15 and 34-37, two As
(Divus Augustus) 15-16 and 22-30, Caius one As 37-38, (Germani-
cus) two As 37-38 and 40-41, Claudius I one As 41-54, Julio-Clau-
dian dynasty one As 27 B.C.-68, Vespasian or Titus one As 69-81,
Traian two As 101-102 and 103-117, Hadrian one S117-138 and one
Fig. 2.12 Dp 117-138, (Sabina) one D 119-137 and one As 119-137.

BEZECZKY_Text.indd 23 18.03.19 14:39

 24 SHORT HISTORY OF THE WORKSHOP AND THE VILLAS

Fig. 2.13 The oil cellar and fragmented dolia in the second villa (Photo: Bezeczky 2014).

Fig. 2.14 Mill stone and a mola olearia (Photo: Bezeczky 2014).

BEZECZKY_Text.indd 24 18.03.19 14:39

 CHAPTER 2 – The villas of Brijuni (La Torre – Bezeczky)
Fig. 2.15 Map of the third Roman villa (Drawing: La Torre 2017).
would an olive oil-producing villa have needed to import oil
from far away? Clearly because local production stalled. Oil
was also imported in amphorae from Baetica under the reign
of Antoninus. The same is true of the African I A/B (110)
and Africana II A (112) amphorae which were transported
to the villa during the mid-second and mid-third centuries.
THE THIRD ROMAN VILLA
It is not known when the third villa was built. Inside the
villa, there was another press-room and a new, small cellar on
the northern wall. As with the other presses, we do not know
their shape, because they are to be found underneath walls
that were built later on. It is clear that the reconstruction was
built during the excavations.
The small cellar is located next to the large one. Its roof
was held up by two columns. The form and material of the
smaller dolia here differ from those in the large cellar. It is
possible that these dolia contained seeds rather than oil or
wine.
BEZECZKY_Text.indd 25
25
There are fewer fine ceramics and amphorae in the villa
from last third of the second century.45 The votive altar stone
dedicated by M. Aurelius Iustus to the goddess Flora in the
cemetery near the villa can be dated to this period.46 It seems
likely that M. Aurelius Iustus used the villa; he may have been
a tenant (conductor) who paid rent to the Imperial treasury.
The Fažana 1 type amphorae that he produced can be
found both in the villa and in the workshop in Fažana. There
is only one amphora with this name M. Aurelius Iustus on the
stamp. Both the amphora and the stamp differ significantly
from the traditions of the workshop. The small amphorae
(Fažana 2) dated to the second or third century pose a mys-
tery. They probably contained fish sauces. It is not inconceiv-
able that salt production took place next to the Castrum villa
(Saline bay), and that the salt was used in the production of
45
According to Girardi-Jurkić 1985, 92, the production of the villa
stagnated for about a hundred years. Pröttel 1996, 165 and 168.
46
ILJug 1204; Mlakar 1979, 23-24; Starac 1995, 138, note 27;
Bezeczky 1998a, 16.
18.03.19 14:39
 26 SHORT HISTORY OF THE WORKSHOP AND THE VILLAS

Fig. 2.16

fish sauces. Few small amphorae have been found on the ter-
ritory of the villa, and there are no basins or buildings related
to fish processing. There are some Cretan fragments of form
MRC 3 and Agora M 125 with micaceous fabric and Agora
M 240 – the fabric without mica – Kapitän 1 and Kapitän 2
amphorae from the Aegean, Africana IIC and Africana IID
dated in the third century.
The circulation of coins was continuous from Antininus
Pius to the Valentinianus III period.47
Regarding the villa, they highlight “Probably in the same
period a fullonica was installed in courtyard of the villa rus-
tica in Madona bay. Among the finds were furnaces, a cistern
and remains of presses for extracting liquid from the fabrics,
spaces for drying fabrics and final processing.”48 The work-
shop was presumably used by an Imperial procurator. It may
perhaps be from a later period, when the state manufacture
of military clothing was connected to the Balfium Cissense
Venetiae et Histriae, mentioned by the Notitia dignitatum
435.49

47
Miškec 2002, 178-187.
48
Suić 1987, 201; Matijasić 1998, 241-251; Begović – Schrunk
Fig. 2.16 The smaller cellar and the dolia (Photo: Bezeczky 2014). 2007a, 98; Tassaux 1998, 96-97.
49
ND Oc I. 53, Oc XI, 67 Procurator balfii Cissensis; Begović –
Schrunk 2007a, 98-99.

BEZECZKY_Text.indd 26 18.03.19 14:39

 CHAPTER 2 – The villas of Brijuni (La Torre – Bezeczky) 27

Fig. 2.17 Map of the Late Roman and Byzantine villa (Drawing: La Torre 2017).

THE LATE ROMAN AND BYZANTINE VILLA OR regarding the layers.51 It is clear that they were used to pro-
SETTLEMENT duce a significant quantity of wine.
In the villa there are very significant amphora finds from
At the time, the villa was surrounded by polygonal walls to
this era. The amphorae come from the Iberian Peninsula,
protect it from attacks. The walls were weakly fortified and
North Africa, Italy, the eastern Mediterranean and the Black
had no towers. Based on the ovens which can still be identi-
Sea.52 Most of the amphorae contained wine, olive oil and
fied, a little more than 30 families lived within the walls in
fish sauce. This raises the question of what the villa pro-
the Byzantine era. The authors refer to historical events when
duced, and why it was necessary to import these commodi-
they describe the late Roman/Byzantine villa. These refer-
ties. Among the late Roman and Byzantine amphorae,53
ences may provide important clues. However, the role of the
there are many Aegean and North African terra sigillata.54
villa can only be guessed at. Starting in the fourth century,
The owners of the villa may have been very wealthy.
the villa was presumably not used as a military site; it may
Or perhaps it was the ecclesiastic leaders of St. Mary’s and
have been closely connected to the neighboring St. Mary’s
St. Peter’s churches who ordered the expensive foodstuffs.
and St. Peter’s churches.50
For details of this period, see the next chapter.
A number of wine presses were found in the villa, but they
are also difficult to date as we do not have any information

51
Matijašić 1993a, 249.
50
The authors of this volume agree that the villa was never a military 52
Vidrih Perko – Pavletić 2000, 266. See the Part II chapters in
base. González Cesteros is an exception. He thinks the villa was used this volume.
by the military from the late Roman period. The excavations have not 53
For the late Imperial and Byzantine finds, see: Girardi-Jurkić
produced any indication that the area was used by military forces in 1985, 93-94, nos. 252-265.
an organized manner. 54
Pröttel 1996, 98, Fig. 15-16, 168, Fig. 71.

BEZECZKY_Text.indd 27 18.03.19 14:39

 28 SHORT HISTORY OF THE WORKSHOP AND THE VILLAS

Fig. 2.18 Wine presses installation (Photo: Bezeczky 2014).

BEZECZKY_Text.indd 28 18.03.19 14:39


3 THE E ARLY CHRISTIAN CHURCHE S OF BRIJUNI
Alexander Schobert
1 THE CHURCH OF ST. MARY
The Church of St. Mary is situated approximately 100 m
north of the fortified settlement (the Castrum). It was built
on top of a small hill, 40 m from Madona Bay, with a view of
Pusti Island to the west. Four separate phases of construction
can be identified on the basis of the preserved architecture.1
The three-aisle-basilica (Fig. 3.1), which was probably
built at the end of the fifth century A.D., was 23.74 m long
and 10.8 m wide.2 The wooden ceiling beams supported a tri-
angular gable without separate mono-pitched roofs for the
side naves. In addition to the main gate on the western side
of the building, the original structure also had two entrances
in the side naves which led directly into the presbytery.
In the interior of the building there are two rows of
monolithic columns dividing the hall into the main and
the side naves. These columns were 2.51 m high and had an
impost without capital to reduce the pressure on the arches.
The imposts (Fig. 3.2) were decorated with Greek crosses
(crux immissa), similarly to Theodoric’s palace in Ravenna.
In the side walls between the columns at a height of 3.8 m
there were windows, measuring 90 × 140 cm. They illumi-
nated the main nave indirectly from the side. Small double
columns decorated with crosses divided the windows, which
were vaulted by semicircular arches (bifora). The sanctuary
faced east, which had no apse, and was lit by a large window
in the back wall.3
Apart from the fact that the Church of St. Mary is some-
what narrower and has only one main entrance at the front,
the layout and the position of the columns are identical to
the Arian Basilica dello Spirito Santo in Ravenna. The latter
was built soon after 493 and represents the first ecclesiastical
building by the Goths in Ravenna. Furthermore, this monu-
1
On the individual phases of construction, see Vitasović 2003,
61-101, Figs. 2-5; the information regarding sizes is drawn from
Begović Dvoržak et al. 2007, 232-238.
2
Bratož 1994, 39-47; Chevalier 1999, 105-117; Ujčić 2005, 91.
3
Begović Dvoržak et al. 2007, 234.
BEZECZKY_Text.indd 29
29
ment indicates the influence of the church of Sant’Andrea
dei Goti in Rome, built 25 years earlier.4
Most of the monolithic platform (1.4 × 0.76 × 0.2 m)
of the early Christian altar has been preserved, lying not far
away from the central position where it stood in the sanctu-
ary (Fig. 3.3). The pedestal has, in addition to the recesses for
the columns in the corners, a small indent (13.5 × 8 cm) for
relics.5 A capital fragment was also found which was used as
the base of the altar plate. Gnirs carried out the first excava-
tions in the Church of St. Mary in 1906-07 and was in favor
of dating it to an early period based on the shape of the altar
platform.6
This hypothesis is confirmed by finds from Ravenna
to North Africa, where this altar type was found from the
fourth-fifth century onwards.7 The fragments of altar parti-
tions with recessed Greek crosses which were found in the
sanctuary also seem to be older than the material of the fifth-
century building.8
The basilica was expanded in the middle of the sixth cen-
tury. A 5.4 m-wide covered portico-like narthex was added
to the western side of the structure.9 An ambo was placed
directly in front of the sanctuary and a cistern (1.8 × 2.77 m)
was installed behind the building.10 The sanctuary was now
separated from the rest of the church by a triumphal arch
4
Bovini 2006, 51-52.
5
Glaser 1997, 244.
6
Gnirs 1911a, 84-86, Fig. 11; this has not been refuted until now, see
Vitasović 2003, 77-84; Begović Dvoržak et al. 2007, 234-235.
7
The classical altar table was prevalent in the west, e.g. in the St. Felicia-
nus Basilica/Pula, Sant’Apollinare Nuovo/Ravenna, but also in North
Africa (Timgad, Tebessa) and Greece (Nea Anhialos, Tassos). See
Braun 1924, 125; Mazzotti 1960, 238-246; Sodini 1984, 17-24,
Figs. 14-23.
8
This type was common in the fourth century A.D. At the end of the
fifth century A.D. one can see that relief crosses were preferred. Com-
parable samples have been found in Strobeč close to Salona and Bijaći
in the Trogir-region. See Migotti 1992, 120; Begović – Schrunk
2007a, 123 Fig. 97.
9
Church architecture flourished again after the Goths withdrew. Some
architectural changes happened in this period, as is evident from the
episcopal complex of Salona. See Bratož 1996, 339-340.
10
Begović Dvoržak et al. 2007, 234-235.
18.03.19 14:39
 30 SHORT HISTORY OF THE WORKSHOP AND THE VILLAS

Fig. 3.1 The Church of St. Mary. A - Narthex, B - Entrance hall built in the main nave, C - Main nave of the
basilica, D - Sanctuary, E - Court, F - Diaconicum, G - Secretarium/sacristy, H - Vaulted corridor,
K - Cistern, L - Socle of the altar, M, N - Place of lecterns, T - Side aisle, U - Block altar of the fifteenth-c.
church, V - Scola cantorum, W - Court (Photo: Pülz 2013, Drawing: La Torre 2017 after Gnirs 1911a).

BEZECZKY_Text.indd 30 18.03.19 14:39

 CHAPTER 3 – The early Christian churches of Brijuni (Schobert) 31

Fig. 3.2 Column of the main nave with impost Fig. 3.3 Monolithic platform of the early Christian altar (Photo: Pülz 2013).
and Greek cross (Photo: Pülz 2013).

Fig. 3.5 Capital in the sanctuary with crux

coronata emblems, Greek crosses and
volutes fitted with a narrow trapezoid
impost stone (Photo: Pülz 2013).
Fig. 3.4 Triumphal arch supported by two monolithic columns with capitals
and imposts (Photo: Pülz 2013).

(Fig. 3.4) supported by two monolithic, 4.2 m-high col-
umns with capitals and imposts which were used as addi-
tional reinforcement for the additional pilasters. The large
triumphal arch was approximately 6 m high. The capitals are
decorated on all four sides with christograms, which were
executed in the middle of the fifth century as crosses in the
center of circular medallions (crux coronata). Thus, a second
use is also obvious for the altar table and screen. There is
proof of modification in the middle of the sixth century: two
columns with bipartite capitals were added to the sanctuary.
They consist of a capital with crux coronata emblems, Greek
crosses and volutes fitted with a narrow trapezoid impost
stone (Fig. 3.5). The same type of capital was used in the
Basilica of Sant’Apollinare in Classe, which was dedicated
in 549. However, just a few years later, it became common
practice to carve capitals with imposts out of a single block
of stone.11
11
Zollt 1998, 62; Bovini 2006, 95-96; Marušić 1967, 33-35;
Begović Dvoržak et al. 2007, 235.

BEZECZKY_Text.indd 31 18.03.19 14:39

 32 SHORT HISTORY OF THE WORKSHOP AND THE VILLAS

Fig. 3.6 Presbytery with septum

(Photo: Pülz 2013).

Fig. 3.7 Hospitium (domus presby-

terorum), northern annex
(Photo: Pülz 2013).

Fig. 3.8 Room with double apses

(cellae memoriae), south-
ern annex (Photo: Pülz
2013).

BEZECZKY_Text.indd 32 18.03.19 14:39

 CHAPTER 3 – The early Christian churches of Brijuni (Schobert)
The expanded presbytery with septum directly in front
of the sanctuary and the construction supporting the super-
structure with arcades in the main nave are from the Justin-
ian Era (Fig. 3.6). These features and the lack of a transept,
which was common at the time, are characteristic of North
African church architecture.12 This is explained by close
trade relations with North Africa and troop movements in
that region until 534. This process of acculturation reflects
the range of the finds in the Byzantine Castrum which, at the
turn of the fifth century, only occasionally includes individ-
ual items from Italy; the eastern Mediterranean and North
African wares, in addition to the amphorae and particularly
African red slip ware dominate.13
The annex on the northern side of the building, next
to the sanctuary, also dates back to the construction phase
around the middle of the sixth century. This annex (13 ×
5 m), which was called a hospitium14 (domus presbytero-
rum), had two floors, a small entrance and narrow windows
(Fig. 3.7).
There was an atrium in front of the building surrounded
by solid walls. On the southern side, in a straight line with
the hospitium, a room (11 × 6 m) with double apses was
added to the nave (Fig. 3.8). A door connected it to the pres-
bytery and another atrium to the south also belonged to it.
12
Three-aisle-basilicas with a raised clerestory were widespread in
Africa. On the main characteristics of North African basilicas, see
Gnirs 1911a, 89-90; Duval 1974, 353-368.
13
This could be due to newly set customs borders, which led to fine
ceramics and foodstuffs from North Africa swamping the western
part of the Mediterranean. For an analysis of the finds, see Bloier
2012, 210 Fig. 11. For more on the finds, see Begović – Schrunk
2006, 191-202; see Bonifay – Capelli in this volume.
14
Begović – Schrunk 2007a, 123-124.
BEZECZKY_Text.indd 33
33
Fig. 3.9 Sarcophagus, narthex
(Photo: Pülz 2013).
The function of this room is unclear. It may well have been an
oratory with alcoves containing relics (cellae memoriae), like
similar spaces with two apses in the early Byzantine churches
of Cilicia and Isauria or the hall with three apses in the Abū
Mīnā Northern Basilica.15
The Church of St. Mary was constantly renovated during
the period of the barbarian invasions. This is clearly reflected
in the architectural decoration, which is similar to chip
carved wood sculpture and to the use of spolia. The continu-
ing decay of the villas near the southern pier is corroborated
by the remains of an architrave portraying a hippocampus,16
which had a secondary use as a foundation for an altar screen.
It was originally part of the temple of Neptune in Verige Bay.
In the course of the third period of reconstruction in the
ninth-tenth centuries A.D., the hospitium was turned into a
dormitorium with a chapter house.17 By building a southern
wing, the Benedictine monks walled up the second entrance
to the Church of St. Mary and increased the width of the
monastery walls by 0.2 m. The initial function of the nar-
thex was lost and it was repurposed as a grave (Fig. 3.9). The
monks entered the complex from the west using the atrium
to the north.18 Marcus Samuelis, Commissarius Briorum was
15
Hill 1996, Fig. 20, 34, 50, 59, 60; Jaritz 1982, 145-148.
16
Gnirs 1911a, 92; Schrunk – Begović 2000, 274; similarly spolia
from Val Catena were used for the Euphrasius Basilica in Parentium.
See Ujčić 2005, 96-98.
17
Although Gnirs 1911a, 90-92 dates the decorative elements of the
hospitium to the seventh century, findings from the oldest known
Croatian Benedictine monastery in Rižinice near Solin, which was
founded by Prince Trpimir in 852, suggest that Mlakar’s dating is
more likely to be correct. See Vitasović 2003, 100.
18
Begović – Schrunk 2007a, 136. The coast north of the castrum had
been used as a necropolis since the Republican period. See Begović –
Schrunk 2007a, 124.
18.03.19 14:39
 34 SHORT HISTORY OF THE WORKSHOP AND THE VILLAS
responsible for the final maintenance work in the fifteenth
century. He was buried in the grounds of the church.19 The
settlement near Dobrika Bay was finally abandoned after the
plague epidemic in 1348.20
2 THE EARLY CHRISTIAN CHURCH WITHIN
THE WALLS OF THE CASTRUM AND IN
VERIGE BAY
It has not yet been possible to verify the existence and loca-
tion of an early Christian church within the Castrum walls.
Gnirs found some architectural fragments with cross decora-
tions in the Castrum as early as 1908.21 The 1976-1983 exca-
vations directed by Mlakar and Vitasović unearthed further
objects from the interior of a church, such as Corinthian
capitals, the remains of pilasters, imposts and barrier frag-
ments.22 However, these items were scattered, which makes
it impossible to specify the exact location of the church on a
map of the settlement.23 The discovery of a lintel with a crux
coronata emblem and tendril motif decorations which had
been reused in a press did not resolve the question either.24
The Church of St. Mary was renovated in the middle of the
sixth century. The stones which were removed were probably
used as spolia.
A polycandelon was found inside a building (20 × 8 m)
reinforced by pilaster-strips on the territory of the Cas-
trum villa.25 It has three holes to accommodate the chain.
Although churches were equipped with similar hanging
lamps, this does not necessarily mean that this building
was in fact a domus ecclesiae. There is also no proof that the
southeastern, circle-shaped avant-corps is what remains of
a church.26 Regardless of the classification of the material,
the fragments of architectural decoration with carved Latin
crosses and broadened ends indicate that they were pro-
duced at an early stage. Such decorations were fashionable
in the Mediterranean region in the 4th and 5th centuries.27
If there was no church in the Castrum, this would suggest
19
Vitasović 2003, 100.
20
Schrunk – Begović 2000, 268-269; Begović Dvoržak 2001,
177-183.
21
Gnirs 1911a, 75-97.
22
Mlakar 1975-76, 10-34; Vitasović 2005, 157-210.
23
See Marušić 1990, 404-413, 405 (sitemap) for an analysis of the
finds.
24
Marušić 1990, 406-408, Figs. 3 and 5.
25
The villa rustica from the early Imperial period (62.8 × 51.2 m) was
u-shaped with an atrium in the middle. It was open to the bay. See
Mlakar 1975-76, 41.
26
Marušić 1990, 405-407.
27
Migotti 1992, 120, 123 Fig. 17 a, b.
BEZECZKY_Text.indd 34
that the Church of St. Mary or a precursor was built earlier,
before the walls of the fortress were erected.
The southern side of Verige (Val Catena) Bay in the east-
ern part of the island seems to be a particularly convenient
location for a church with a baptistery. Gnirs28 was the first
to unearth parts of a luxurious maritime terrace villa,29 which
formed the imposing center of the latifundium.
A temenos with temples for Venus, Neptune and Mars (?)
close to the building was erected in the west (see the previous
chapter Fig. 2.2, M-N-P and E-F-G-H). The complex also
contained a library, a balneum with a portico, an industrial
wing with equipment for oil production, piscinae for fish
production and a pier for docking to the northeast.30
The thermae of the port were probably repurposed as an
early Christian basilica during the reign of Theodosius, when
Christianity became the religion of the empire. The tepidar-
ium with an apse could have been rededicated as a nave. The
1.2 m-deep pool lined with white marble in the former frigi-
darium seems to be used as a baptismal font.31 In 1903, Gnirs
found capitals decorated with dolphins in this room; they
were originally part of the temple of Neptune. Early Christian
graves were found outside in the vicinity of the apse, which
may confirm that the building was repurposed as a church.32
The transformation of the port thermae and the con-
tinued use of the olive oil presses illustrate the interaction
between the two bays. However, the hub of the island ulti-
mately shifted from the villa complex in Val Catena to the
western side of the island by the beginning of the fifth cen-
tury at the latest.33
3 THE CHURCH OF ST. PETER
An observation post with a Castrum and a church was con-
structed on the Petrovac hill (33 m above sea level), approx-
imately 170 m northeast of the Church of St. Mary, in an
area which is now densely wooded.34 The advantage of this
location was that it enabled observers to get early warning of
enemy ships and thus warn the port settlement in time.
28
Gnirs 1904, 145-146.
29
On the villa maritima in Val Catena: Lafon 2001, 459; Schrunk –
Begović 2000, 257-261; see La Torre – Bezeczky in this volume.
30
In 1996-97 a Bavarian/Croatian team carried out underwater research
on the northern pier. On trade goods and their analysis, see Bloier
2012, 197-205.
31
Begović – Schrunk 2007a, 113; Džin 2011, 97.
32
Gnirs 1915, 137, 141.
33
On the statistics of the finds see Bloier 2012, 196, 210, 227.
34
Begović Dvoržak 2001, 183-187 for a detailed description of the
structure.
18.03.19 14:39
 CHAPTER 3 – The early Christian churches of Brijuni (Schobert)
Parts of the excavated defensive structure form a half-cir-
cle complemented by a small buttressed fortress. The fortifi-
cation and the rest of the structure form a single architectural
unit. A rectangular building (4.6 × 5.5 m) is visible directly
in front of the church; its walls are up to three-quarters of
a meter high. The entrance to the tower, which presumably
served as an observation deck, was on the southern side.35
The hall-church has a rectangular layout (5.82 × 7.78 m).
Its front area slopes to the south. The preserved bottom part
of the main entrance (2.05 × 2.03 m) together with the sur-
rounding stonewalls makes a precise reconstruction possible
(Fig. 3.10).36 A polygonal apse formed the outer end of the
Church of St. Peter. On the inside, it was horseshoe-shaped
with a 2.9 m diameter.37 It contributed to the structural bal-
ance of the gabled roof to the northeast together with the
triumphal arch in the center of the spatial axis and the half-
meter-wide pilaster-strips of the outer walls.38 This type of
construction reflects Ravenna architecture and proves that
the church was built during the same period as the Claustra
Alpium Iuliarum.39 The Church of St. Nicholas in Pula from
the fifth/sixth century has a similar structure and dimensions.40
35
Begović Dvoržak 2001, 186.
36
Gnirs 1911c, 41-44.
37
Unlike the semicircular apse, which was already used in the secular
architecture of the Imperial period, polygonal apses make their first
appearance in the beginning of the fifth century in urban centers. See
Migotti 1992, 118-119.
38
Mohorovičić 1957, 491-493.
39
The “Adriatic limes” was composed of an unbroken line of walls and
guard posts and served primarily to protect transportation routes, spe-
cifically to protect trade routes in the time of Justinian. See Šašel –
Petru 1971.
40
Rudolf Egger had already pointed out that this type of church had
been established relatively early, especially in the Alps-Adriatic region.
See Egger 1916, 122; Chevalier 1999, 108.
BEZECZKY_Text.indd 35
35
Fig. 3.10 The Church of St. Peter,
main entrance (Photo:
Bezeczky 2013; Drawing:
La Torre 2017, after Bego-
vić – Schrunk 2007a).
The Church of St. Peter had another entrance to the
north, which gave the clergy direct access to the presbytery.
The sanctuary is slightly raised and separated from the hall
by six altar partitions. The traverse under the floor, which
had clearly recognizable hollows to hold the septum, was
preserved along its entire length. Altar screen remains used
as spolia can also be found in situ. The limestone slabs (75 ×
85 × 7 cm) are decorated with Greek crosses surrounded
by zig-zagging lines and an interlaced pattern. Based on the
Carolingian style of the altar screens, renovations took place
in the seventh/eighth century. Gnirs found some items with
relief ornaments from the early Christian period, covered by
the fragments of a triumphal arch, during his inspection of
the area outside the church in 1910.41
There is a partially preserved polychromatic floor mosaic
in the church. It is a 0.35 m-wide strip with red, white and
blue tesserae laid out in a rectangular pattern. Intersecting
circles containing Greek crosses are connected by wavy lines.42
Water was provided using a trapezoid cistern (4.5 × 2.6 m)
on the northern side of the building. Walls were discovered a
few steps away which served to further separate the temenos.
This suggests that there was a civilian settlement nearby, as
the church complex would not have needed more fortifica-
tions otherwise. The large number of tegulae fragments and
clearly identifiable, unbroken walls on the northern side of
the hill may corroborate this hypothesis.43
41
Gnirs 1911c, 42; Begović Dvoržak 2001, 184.
42
Begović Dvoržak 2001, 185. A similar polychrome mosaic was
laid out at the beginning of the fifth century in the ambulatory
between the apse and the synthronon in the Episcopal Basilica at
Zadar. See Vežić 2005, 38-41. On an overview for mosaics in Adri-
atic church centres, see Campanati 2008, 435-454.
43
This hypothesis is based on Marušić’s inspection of the site. He found
the scattered remains of many roof tiles and wall fragments there. See
Begović Dvoržak 2001, 186 and see Berni Millet in this volume.
18.03.19 14:39
 36 SHORT HISTORY OF THE WORKSHOP AND THE VILLAS

4 THE BISHOP OF CISSA PULLARIA (CESSUS)

AND HIS RESIDENCE
The island of Brijuni was called Cissa Pullaria in Antiq-
uity, as one sees in the writings of Pliny the Elder44 and the
Peutinger Table (Fig. 3.11).45
The island had a special administrative status going back
to the Flavian period, and it was under direct Imperial con-
trol.46 This is corroborated by the inheritance of the land-
owners Laecanii. They owned the latifundia on Brijuni
Island. When the last Laecanius died without an heir, all of
his property came into Vespasian’s ownership.47
Fig. 3.11 Peutinger Table pars V.
After the middle of the second century A.D., military
pressure on Rome increased constantly. To understand the
the sea. It was rectangular in shape, and covered one hectare
whole situation in respect of social and economic changes,
along the coast.51
the archeological finds from Brijuni had to be compared
The name of the island was altered along with its legal sta-
with material from Pula, but a complete publication about
tus, and it is referred to as Census (Cessus) in records from
these amphorae is still pending.48
Late Antiquity.52
Geopolitical and cultural changes took place in the
Brijuni formed the important northern endpoint of the
course of the fourth-fifth centuries A.D. following the battle
defensive system which followed the maritime route to
of Adrianople in 378. There were changes in the power rela-
Ravenna along the eastern Adriatic coast. This was signifi-
tions in the Adriatic region following the Huns’ destruction
cant, as these maritime routes were extremely dangerous in
of Aquileia (452) and the end of Western Roman Empire in
the sixth century. The island was also home to a military
476.49 The first defensive wall was constructed around the
outpost which protected the episcopal centers of Pola and
civilian settlement which sprang up around the villa rustica
Parentium/Poreč.53
in Dobrika Bay in the fourth century. In the time when Bri-
The secular power of the Church increased during the
juni was ruled by the Ostrogoths, Cassiodorus, senator and
reign of Justinian I, and Brijuni must have had a bishop in
praetorian prefect of the province of Istria (533-537), men-
this period. This is supported by a deed of gift dating from
tioned the island in a letter. He noted that it was an impor-
534 A.D. in which Bishop Euphrasius of Parentium donated
tant source of supplies for the capital Ravenna, comparable
one-third of the salt production from Val Laura (in the south-
to the latifundia of Campania.50
ern part of the island) to the church of Brijuni.54 Documents
The fortification in Dobrika Bay (Castrum) was probably
from the Synod of Grado in 579 contain the first written
constructed when Iustinian I began to reconquer Istria at the
record of an “Episcopus Cessensis”, referred to there as Vinde-
latest. The walls were 2.6 m wide. They were equipped with
mius.55 Vindemius was inclined towards the Lombards in the
balustrades and battlements. The walls were approximately
Three-Chapter Controversy. The Lombards followed Arian-
5 m high, and large cut stone blocks were placed at the ends
ism, which meant that they doubted the consubstantiality
to add stability. The structure had five gates with little tow-
of the Father and the Son, and found the discussions regard-
ers on top for extra protection on the side facing away from

51
Gnirs 1911a, 94-95; Begović Dvoržak et al. 2007, 232;
Matijašić 2012, 193-208.
44
Pliny NH 3, 151, 152. 52
Suić 1987, 201-213; Vicelja Matijašić, 2007, 48.
45
Križman 1979, 314-316, 321. 53
Many sixth century fortifications have been preserved in Croatia
46
Brijuni only came under the control of the administration in Pula in and Slovenia, which formed part of the Justinianian defense system
Late Antiquity. See Suić 1987, 201-213. (Claustra Alpium Iuliarum). They were erected at strategically impor-
47
Bezeczky 1998a, 15 and 68; Bezeczky 2014 and see also his article tant locations inland as well as on islands along the Adriatic coast. See
in Chapter 1. Poulter 2013, 63-76; Ciglenečki 2009, 205-222; on the castrum
48
On the eastern Mediterranean amphorae, see González Cesteros in see Mlakar 1975-76, 10-34; Begović – Schrunk 2009, 223-236;
this volume. Begović – Schrunk 2011a, 375-390.
49
Matijašić 2012, 150; Bratož 1996, 331-351. 54
Schrunk – Begović 2000, 268.
50
Cassiod. var. 12, 22; see Begović – Schrunk 2007a, 103-104. 55
Begović – Schrunk 2011b, 658.

BEZECZKY_Text.indd 36 18.03.19 14:39

 CHAPTER 3 – The early Christian churches of Brijuni (Schobert)
ing the relationship between the divine and human nature
of Christ absurd.56 The bishop was arrested in 588 in Raven-
na.57 During his year-long incarceration, he had enough time
to reconsider his opinion on Monophysitism. As a partici-
pant at the Synod of Marano in 590, Vindemius had already
adopted the Catholic faith.58 There are records of another
Bishop Cenetensis in the person of Ursinus, who partici-
pated in the sixth ecumenical council in 680 in Rome.59 This
confirms that an episcopal center existed in Brijuni from the
middle of the sixth to the end of the seventh century. The
bishop’s seat was moved to Ruginium/Rovinj around 800.60
Before an independent diocese was established in Brijuni,
the visiting bishop could have used the church in Verige
Bay for baptisms (see chapter 3, Fig. 2.2, F). The villa mari-
tima was an obvious choice for the episcopal residence. At
the turn of the fourth-fifth centuries, when the settlement
in Dobrika Bay became the main settlement of the island,
either a little church with a residence attached was built in
castellis or already the Church of St. Peter.61 It may have been
the original location of the altar of the Church of St. Mary,
as it is older than the preserved construction materials. The
56
Bratož 2000, 640-641.
57
In his Historia Langobardorum (3, 26), Paulus Diaconus claims that
the Patriarch had the schismatic Videmius ex Istria arrested. See
Bratož 2000, 64. Suić 1987, 207 clings to his theory that Vide-
mius was arrested in the bishop’s residence in Brijuni.
58
Marušić 1990, 403.
BEZECZKY_Text.indd 37
37
construction of the Church of St. Mary commenced when
the Episcopus Cessensis was appointed, and as the majority
of the parish had already received the sacrament of baptism,
there was no need to set up a baptistery. For newly converted
Christians, children or migrants, baptism by aspersion was a
possible alternative.
After the hospitium was completed, the bishop moved
to the annex of the Church of St. Mary or used the terrace
house on the Petrovac hill, which Gnirs described as a “mon-
umental building with Ionic capitals” in 1911.62 The area has
not been examined again since. However, the terrace house
is the most probable location for the episcopal residence.
On the one hand, it is located in between the churches of
St. Peter and St. Mary, and on the other, the bishop would
have been able to find safety quickly in the fortress in the
event of an attack. The Castra provided shelter for the popu-
lation, but they were not a military fortification. This is also
demonstrated by the location of the two churches outside of
the walls. The evidence presented here indicates that there
was undoubtedly a bishop’s see in Brijuni which existed for
more than two and a half centuries.
59
Schrunk – Begović 2011a, 680; for more on the synod see
Bratož 1998, 589-602.
60
Bratož 2000, 643-644.
61
See Begović – Schrunk 2010, 263 for a differing view.
62
Gnirs 1911a, 93.
18.03.19 14:39
BEZECZKY_Text.indd 38 18.03.19 14:39
 THE A MPHOR AE

BEZECZKY_Text.indd 39 18.03.19 14:39

BEZECZKY_Text.indd 40 18.03.19 14:39

4 ITALIAN AND ISTRIAN A MPHOR AE
Tamás Bezeczky
Amphorae produced in Italy and Istria make up 27 % of all
amphorae excavated at the Castrum villa. We can differenti-
ate between two types, those made locally and those trans-
ported to the villa from other parts of Italy.
To review Istria’s economic history, one must examine
many Adriatic amphora types: Lamboglia 2, Dressel 6A,
Dressel 2-4, Dressel 6B, Fažana 1, Fažana 2, Loron small
amphora, Porto Recanati / amphora con collo ad imbuto,
Schörgendorfer 558, flat based / Forlimpopoli, north Adri-
atic fish sauces amphorae, Aquincum 78, amphora from
Grado and Bónis 31/5. A significant number of these
amphora types were also found at the Castrum villa. Many
studies examine where they were produced and what they
contained, but current answers to these questions are not
always convincing. Clear typological and petrographical
descriptions are needed. The amphorae published from
northern Italy do not always match the finds from Nori-
cum and Pannonia. The recently published amphorae with
stamps and painted inscriptions will be very helpful in deter-
mining their origins.1
1 ITALIAN WINE AMPHORAE FROM THE
LATE REPUBLICAN AND EARLY ROMAN
PERIOD
LAMBOGLIA 2
The earliest dated amphorae found at the Castrum villa
belong to the Lamboglia 2 type. Lamboglia 2, one of the
best-known amphorae from the Adriatic coast, was widely
used across the Mediterranean region. It is a robust vessel
with thick walls. In fragmented form it is difficult to distin-
guish it from the type of amphora that later came to replace
it, the Dressel 6A. The two amphora types are generally
made of the same fabric, as the Dressel 6A amphorae were
made at certain Lamboglia 2 workshops. There has been a
1
The papers of the volume eds. Pesavento Mattioli – Carre
2009.
BEZECZKY_Text.indd 41
41
great deal of research on this amphora since the publica-
tion of Nino Lamboglia’s article.2 Based on the published
data, these vessels were produced in a number of places from
Aquileia to Brindisi.3 Central Dalmatia was also recently
mentioned as a production area, thus starting a lively dis-
cussion.4 It is necessary to examine, as soon as possible, the
Adriatic amphora types mentioned in the introduction, the
amphorae produced in known or presumed workshops, and
geological samples from the surrounding area.5 The pro-
duction of Lamboglia 2 amphorae began between the end
of the second century B.C. and the emergence of the Dres-
sel 6A amphorae in the final third of the first century B.C.6
Although it is impossible to specify when exactly Dressel 6A
production began, it is likely to have been in the last third of
the first century B.C.7 There are two stamps on the Castrum
amphorae. The BAR·ME(…) stamp (2) has not been found
elsewhere. The other Castrum stamp (3), VERSO, is widely
2
Lamboglia 1955, 262.
3
Panella 1970, 117; Will 1970, 383-386; Tchernia 1986, 68-74;
Empereur – Hesnard 1987, 33; Cipriano – Carre 1989, 85;
Cipriano 1994, 207 Fig. 3; Sciallano – Sibella 1994; Bruno
1995, 83-92; Zupančić et al. 1998, 345-357; Monsieur et al. 2006,
310-313; recently Rizzo 2014, 120-123; Carre et al. 2014, 417-428
with complete bibliography.
4
Cambi 1989, 321; Lindhagen 2009; Lindhagen 2013; Kirigin
et al. 2006, 193-194; Panella 2011, 96-97; Carre et al. 2014, 417-
428.
5
It would be helpful to use the petrographical analysis of the
Lamboglia 2 amphorae found in Ephesus for further research.
Bezeczky 2013a, 114-120, see Roman Sauer’s results: as previously
known, Lamboglia 2 amphorae were produced in the workshops
excavated near Brindisi (Apani and Giancola). This was proved using
the petrological analysis of the amphorae found in Ephesus; fabrics
P and P1 are characteristic of the area around Brindisi. Another
fabric group (O) indicates a central Italian origin. Finally, the marine
clay used for fabric group P2 is also similar to the various types of
amphorae made in Italy. See also Sauer 2013, 120-124.
6
Tchernia 1986, 53-56; Grace 1965, 11; Baldacci 1972, 109;
Van der Werff 1986, 103; Peacock – Williams 1986, 100;
Carre 1985, 211.
7
Carre 1985, 211; Van der Werff 1986, 103; Peacock –
Williams 1986, 100; Tchernia 1986, 134.
18.03.19 14:39
 42 THE AMPHORAE
known.8 The content of these amphorae was wine.9 A sig-
nificant quantity of Lamboglia 2 amphorae was found at the
excavation near Pula harbour.10 This is also one of the best-
known amphora types in Aquileia and in the Trieste area.11
The pieces transported from Aquileia are primarily found in
Magdalensberg (Noricum), but they have been found occa-
sionally in the Drava-Sava valley, conquered by Augustus.12
The form is common in the western Mediterranean, present
in Greek regions and in North Africa.13
DRESSEL 6A
Dressel 6A amphorae differ only slightly from the Lam-
boglia 2. The design of the handles is somewhat different,
and the stubs of the Dressel 6A amphorae are pointed.14
The production of these amphorae began in the final third
of the first century B.C. and continued until the middle
of the first century A.D. This is corroborated by the pieces
found in the Carthage amphora wall, as well as the painted
amphora inscriptions published from Rome and Magdalens-
berg which include the dates of consuls.15 Amphorae from
the southern and western areas of Pannonia are a form of
indirect proof that has barely been taken into account. The
Dressel 6A amphorae are found in the military camps and
settlements along the stretch of the Amber Road conquered
by Tiberius (Siscia, Sirmium, Gomolava, Emona, Salla,
8
Desy 1989; Bruno 1995; Carre et al. 1995; Gabucci – Quiri
2008, 65; Brecciaroli Taborelli 1987, Pl. XLI 28.3.
9
Formenti et al. 1978, 95-100; Tchernia 1986, 53; Empereur –
Hesnard 1987, 33.
10
Starac 2008, 123-168.
11
Carre – Cipriano 1985, 6-23; Cipriano – Carre 1989, 97-98;
Verzár-Bas 1991, 182-194; Verzár-Bas 1994; Horvat –
Bavdek 2009, 84-93; Auriemma et al. 2008, 171-173.
12
Maier-Maidl 1992, 95-103; Bezeczky 1994, 15; Bezeczky
1995, 156.
13
Tchernia 1986, 53-56, 69-74; Beltrán, 1970, 349-358; Grace,
1979, Fig. 36; Will 1987, 202-204; Will 1989b, 302-305, Figs.
7-11; Cipriano – Carre 1989b, Figs. 14, 97-99; Martin-
Kilcher 1993, 277-279; Martin-Kilcher 2005, 208, 212;
Bruno 1995, 83-275; Župančić et al. 1998, 345-357; Ehmig 2003,
46; Lindhagen 2009, 95, Fig. 4; Maier-Maidl 1992, 95-103;
Lund 2000a, 84; Monsieur 2001, 75, Fig. 12; Bezeczky 2005a,
Lamboglia 2; Bezeczky 2013a, 114-120; Vidrih Perko 2006,
211; Auriemma 2006, 168-169; Mazzocchin 2013, 66-67;
Rizzo 2014, 122-123; Carre et al. 2014, 417-428.
14
Baldacci 1967-68, 7-49; Buchi 1973, 531-637; Carre 1985,
211; Böttger 1992, Fig. 1. nos. 3-5, T. 96, no. 3 T. 97; Martin-
Kilcher 1993, 277-279.
15
Delattre 1894, 89-110 = CIL, VIII. 3, 2200, dated between
43 and 15 B.C.; CIL, XV. 4582, A.D. 36; Carre 1985, 211-213;
Piccottini 1997, 204.
BEZECZKY_Text.indd 42
Savaria, Devin).16 No Dressel 6A amphorae have been found
in eastern Pannonia, conquered in the late Claudian period.
We know of only one stamp in the Castrum villa, con-
sisting of the letter V stamped on a rim (5). It has not been
found elsewhere to date. A post cocturam graffito of the num-
ber XXIIII can be found on the rim of another amphora
(7). The questions regarding the workshops where this
amphora type was produced are similar to those regarding
the Lamboglia 2 amphorae.17 It is important to note that
these amphorae were not produced on the Istrian peninsula,
even though some recent publications consider them to be
Istrian.18 The inscriptions on the Dressel 6A amphorae dis-
covered in Rome and published by Dressel mention wine,19
although the inscriptions of some amphorae from Salzburg
and Italian sites refer to fish sauce.20 However, in all probabil-
ity, these amphorae were generally used to store wine. They
are fairly common in Aquileia, Noricum (Magdalensberg)
and western Pannonia.21 They also reached Italy and the
western Mediterranean, North Africa, Greece, the eastern
Mediterranean, the Red Sea and India in fewer numbers.22
2 OLIVE OIL AMPHORAE
The olive oil produced in Istria is well-known. It was praised
by ancient sources, which considered it equal in quality to
olive oil from Baetica. Olive oil was also produced in north-
ern Italy, and transported using Dressel 6B amphorae.23
16
There are a few pieces in the Claudian layers in Ostia, but they are
missing from the Flavian layers: Carre 1985, 213; Bezeczky 1994,
130; Bezeczky 2013, 120.
17
See note above nos. 1 and 2.
18
Zevi 1967; Buchi 1973, 547; Will 1996, 208. It seems to be a
misunderstanding to suppose that the Istrian amphorae can be found
in Athens: Will 1997, 124. This is based on improper identification
of the two types (Dressel 6A and 6B).
19
CIL, XV. 4582. 4583. 4653. 4852; Zevi 1966, 217.
20
Heger 1986, 132-135, Fig. 1; Carre et al. 2009, 216-217.
21
Carre 1985, 209-218; Cipriano – Carre 1989, 85-88; Verzár-
Bas 1991, 195-196; Verzár-Bas 1994; Maier-Maidl 1992, 80-95;
Bezeczky 1994, 22-34; Bezeczky 1995, 156.
22
Baldacci 1967-68, 27-29; Riley 1979, 151-156; Carre 1985, 211-
213; Tchernia 1986, 129-133; Cipriano – Carre 1989, 85-88,
Figs. 17, 100; Maier-Maidl 1992, 86; Pesavento Mattioli –
Zanini 1993, 47, Figs. 49-51; Martin-Kilcher 1993, 277-279;
Bezeczky 1995, 156; Bezeczky 1998b, 228-230; Bezeczky
2013, 120-122; Bruno 1997, 517; Martin-Kilcher 2005, 208
and 212; Lund 2000a, 84; Rizzo 2003, 149; Lindhagen 2009,
96; Ehmig 2003, 47; Tomber 2005, 230; Vidrih Perko 2006,
211; Auriemma 2006, 168-169; Auriemma et al. 2008, 171-173;
Carre et al. 2014, 417-428; Rizzo 2014, 123.
23
Some painted inscriptions in Magdalensberg mention Istrian oil,
Piccottini 2000-01, 378; most recently, Carre et al. 2009, 217-
18.03.19 14:39
 CHAPTER 4 – Italian and Istrian amphorae (Bezeczky) 43

DRESSEL 6B (1A PHASE), ANTE DRESSEL 6B rae from northern Italy, but is also similar to the amphorae
produced in the workshop at Fažana.
Few amphorae of the earliest Dressel 6B (1a phase) type have
been found at the Castrum villa. These are primarily rim and
base fragments. This is a transitional type, which is similar DRESSEL 6B (2A–3A PHASE), CLASSICAL
to both the Brindisi / Adriatic ovoid type and the classic DRESSEL 6B
Dressel 6B. These amphorae were termed ante Dressel 6B
The Dressel 6B (2a–3a phase) amphorae from Fažana are
by Alessandra Toniolo.24 Marie-Brigitte Carre and Stefania
one of the best-known products of the Istrian peninsula.30
Pesavento Mattioli later provided a more precise description
The workshop clearly produced amphorae in large quanti-
of the different variants of the Dressel 6B form, distinguish-
ties, with many potters collaborating, which lead to differ-
ing between production in Cisalpine Gaul and Istria.25 They
ences between the pieces belonging to individual batches.
used late Republican stamps of people who held office at the
We know the rims best, as these were preserved because
end of the Republican period to date the ante Dressel 6B
of the stamps. The few entire amphorae which have been
type.26 The distribution of the ante 6B outside northern Italy
published reveal the differences quite clearly: there are dif-
was limited, and even in the northern region they have only
ferences even between items that bear the same stamps.
been found at Magdalensberg.27
Few Laecanius and Imperial amphorae have been found
The Laecanii workshop in Fažana also produced Dressel
at the Castrum villa. Regarding the Laecanius amphorae,
6B - ante 6B (phase 1A) amphorae. Such an amphora, bear-
the LAEK·A (11), [C·LAEK] / SYNT (12), [LAEK] /
ing the stamp C·LAECANI·P·F, was found in the Venice
COMVS (13), LAEK (14), stamps belong to the Tiberian–
lagoon. This stamp matches a tegula stamp from the Cas-
early Claudian (Magdalensberg II, A.D. 14-45/50) group.31
trum villa.28 The ante 6B type amphorae are important proof
Among the stamps which belong to the Claudian–Flavian
of the beginning of production in Fažana, invalidating all
period (Post Magdalensberg I, A.D. 45/50-78), CL[AEK]
previous opinions, which dated the first amphorae from the
(15) matches the stamps found in Celeia and Poetovio.32
Laecanii workshop to the Tiberian period.29
Only the letters E and K written in ligature have survived
Petrological analyses suggest that the ante Dressel 6B
on one stamp, which can be grouped with several different
amphorae (9-10) found in the Castrum villa were not pro-
stamps [CLA]EK (16). The PA[GANI] stamp (17) can
duced in the workshop at Fažana or, if they were produced
be dated to post Magdalensberg II (A.D. 45/50-78) or the
there, must have been the workshop’s earliest products. Dif-
Vespasian (A.D. 78-79) period. The IM[P] stamp (18) also
ferent raw materials were used to produce these amphorae
belongs to this group.33 An amphora with a particularly dif-
and the amphorae bearing stamps from Italy (AP·PVLCRI,
ficult to decipher stamp, LESBI (22), was produced in the
SEPVLLI P·F, P·Q·SCAPVLAE). The analysis of the Cas-
Domitian period (A.D. 81-96).34 Three amphora rims with
trum (9) amphora reveals that the composition of the parti-
the IMP·TRA stamp were also found, and can be dated to
cles in its material is somewhat similar to that of the ampho-
Trajan’s reign (A.D. 98-117). The stamp of the workshop
manager appears next to the IMP·TRA stamp (19) on one
of the rims. Unfortunately, it is hard to decipher, as only the
218 mentioned that the Dressel 6B amphorae occasionally contained last two letters of the name or its abbreviation are visible …
other things, based on the painted inscriptions. VS.35 The final Dressel 6B amphora was produced during the
24
Toniolo 1991, 21-23.
25
Carre – Pesavento Mattioli 2003, 453-476.
26
Appius Claudius Pulcher was consul in 38 B.C.; one of the members 30
Carre – Pesavento Mattioli 2003, 462.
of the Sepullius family, P. Sepullius Macer was quattuorvir monetalis 31
Bezeczky 1998a, nos. 1, 507 and 611; Vidrih Perko – Pavletić
in 44 B.C. The Sepullius families are known from inscriptions in 2000, 265.
Padova; Buchi 1973, 585-588; Carre – Cipriano 1985, 12; 32
Bezeczky 1998a, nos. 37-38.
Zaccaria 1989, 481; Tassaux 2001, 502-503; Toniolo 1991, 33
Bezeczky 1998a, no. 648; Vidrih Perko – Pavletić 2000, 265.
168; Cipriano – Mazzocchin 2000, 169-184; Cipriano – 34
Siscia inv. no. R 7935; Concordia, Cipriano 2008, 304-305, Fig. 1,
Mazzocchin 2002, 307-311. [IMP]·DOM / LESBI.
27
Bezeczky 1998b, 235. 35
There is one amphora in Vindobona with two stamps: IMP(eratoris)
28
Toniolo 2011, 189-192; Matijasić 1987, 164, no. 25; Matijasić [Nervae] TRA(iani) (upside down, M-P letters in ligature) and
1998a, 253. Unfortunately, we were unable to carry out petrological SERV(...) with the S written backwards. The abbreviation SERV(...)
analysis, as we did not have samples of either the amphora or the can be interpreted in various ways: Servacus, …. Serus, or simply
tegula. Serv(us), Bezeczky 2005b, 48. The Castrum workshop manager’s
29
Martin-Kilcher 2000, 508. stamp is different from the previous one.

BEZECZKY_Text.indd 43 18.03.19 14:39

 44 THE AMPHORAE
reign of Hadrian. The amphora finds prove that the villa was
in use until Hadrian’s reign. During this period, the Istrian
producers were unable to export enough olive oil to their tra-
ditional markets, perhaps because of bad weather – some of
the trees probably froze – or some other unknown reason.36
From then on, the provinces of Raetia, Noricum and Panno-
nia imported olive oil from Baetica, transported in Dressel
20 amphorae. A stamp from a Dressel 20 amphora (40) has
been found in Castrum villa. It was probably on an amphora
used to transport olive oil from Baetica, to make up for the
lack of local olive oil.37
FAŽANA 1, ISTRIAN DRESSEL 6B (4A PHASE)
As already mentioned, the workshop in Fažana started pro-
ducing Fažana 1 (Dressel 6B 4a phase) amphorae bearing the
stamp of M. Aurelius Iustus in the last third of the second
century A.D. With the appearance of this amphora, it is as
though the form of the Dressel 6B had been forgotten. It had
a short, straight rim, reinforced at the neck (26). The classic
6B form had a stub, while this form had a pointed spike. The
whole amphora is rougher. While these amphorae initially
appeared to be the latest variation on the classic 6B form,
it now seems that they belong to a type of their own, which
archaeologists call type Fažana 1.
The stamp of M. Aurelius Iustus is very different from
the stamps of the Laecanian and Imperial amphorae. This is
a stamp in hollow lettering, in the genitive case. Only one
similar stamp was found during the new Fažana excavation:
F♠A with a pattern of leaves.38 The quantity of the Fažana
1 amphorae found in the Castrum villa is one-tenth that of
the Dressel 6B amphorae. However, they were produced for
a much shorter period of time, and to date have only been
found on the Istrian peninsula. M. Aurelius Iustus probably
attempted to reconquer the lost markets.39 He was unsuc-
cessful, and his amphorae can only be found in Istria.40
36
Bezeczky 1998a, 10.
37
See in the catalogue no. 40 and also see González Cesteros – Berni
Millet in this volume.
38
Paić – Bulić 2008, 28, no. 34.
39
Bezeczky 1998a, 16; Tassaux 1982, 254 he dated to the Laecanius
period; later he (Tassaux 1998, 93) dated it to the end of the second
century.
40
Vidrih Perko – Župančić 2011, 154.
BEZECZKY_Text.indd 44
3 ISTRIAN FISH SAUCES AMPHORAE
FAŽANA 2, ISTRIAN DRESSEL 6B (4A PHASE)
A small amphora which differs from all the other types pro-
duced in the workshop was given the name Fažana 2 (Dres-
sel 6B 4a phase). The rim is chalice-shaped and the handle is
round in section (27). Gnirs described these amphorae after
excavating the workshop, and some fragments can be seen
in the Brijuni Museum.41 The rescue excavation of 1990-
1991 unearthed rim and handle fragments which belonged
to these small amphorae.42 For the moment, there is not
enough information to date the Fažana 2 amphorae, and we
must wait for the detailed publication of the excavations.
The clay used to produce Fažana 1 and 2 amphorae is similar
to that used for the Laecanius and Imperial amphorae.43
This form is probably connected to the Bónis 31/5 ampho-
ra. Éva Bónis found this 32 cm-high amphora in the Emona
44
cemetery (in grave no. 934, Grabfeld Wienerstrasse). She
dated the finds and the amphora to the first-second century
A.D. Sonja Petru later published the necropolis of Emona
and, taking the rest of the finds into consideration, dated the
amphora to the second century A.D.45 The amphora found
in Iseum in Savaria can also be dated to the second century.46
The other small amphora published by Bónis is also from
Emona, but from a different cemetery (Grabfeld Lenarčič).
Based on the photograph, this amphora is most similar to
the amphorae produced in the Loron workshop.47 Found in
Pannonia, in the collapsed cellar of a villa in Baláca, it also
belongs to this group. The demolished layer was dated to the
end of the second century or the beginning of the third cen-
tury.48 This suggests that the workshops in both Fažana and
Loron started producing such amphorae around the middle
or end of the second century.49
41
Gnirs 1910a, 81, no. 4; Bezeczky 1998a, 9, Fig. 8.
42
Bezeczky – Pavletić 1996, 157, Pl. 3, nos. 9-10.
43
Bezeczky – Mange 2009, 253.
44
Bónis 1942, 232, T 31, no. 5, inv. no. 6397, grave no. 934.
45
Petru 1972, 88.
46
Bezeczky 1987, 78, no. 320, Fig. 30. Based on the thin section,
Szakmány and Józsa are of the opinion that all of the inclusions are
similar to the Fažana amphorae but opalic spiculae are missing.
47
Bónis 1942, 232, T 33, no. 7, inv. no. 6764, grave no. 70; Marion –
Starac 2001, 117; Marion 2009, 284.
48
Palágyi 2007, 287.
49
It is possible to determine where the individual pieces were produced
based on the petrological analyses. For new evidence of other
workshops see Vidrih Perko – Župančić 2011, 154.
18.03.19 14:39
 CHAPTER 4 – Italian and Istrian amphorae (Bezeczky)
4 NORTHERN ADRIATIC AMPHORAE
AQUINCUM 78
There are rim fragments of the Aquincum 78 type amphora
in the Castrum (32-33). One such amphora was found in
1978 in Aquincum (Pannonia).50 The formal characteristics
of the amphora resemble the Dressel 6B type. The body is
oval-shaped ending in a short stub. The thin, small handles’
cross-section is circular, buff in colour (10 YR 8/3). The
dimensions of the Aquincum amphora are the following:
diameter of the rim 10 cm, height 77 cm. Chronologically
it was used between the end of the first to the end of the sec-
ond century A.D. It is widely distributed throughout Panno-
nia (Poetovio, Salla, Arrabona, Mursella, Tokod, Aquincum,
Mursa). Its place of production is currently unknown.
5 MISCELLANEOUS ITALIAN AMPHORAE
PORTO RECANATI / ANFORA CON COLLO AD
IMBUTO
A few Porto Recanati amphora fragments were found in the
Castrum villa (30-31). The rim and the neck of this Porto
Recanati type are funnel-shaped.51 Some of them have a sin-
gle line on the rim, others have a horizontal groove. The finds
in Italy make it possible to identify two sub-types.52 Few
stamps are known on this type of amphora.53 It is certain
that they were not produced in the Fažana workshop. Pet-
rological analyses showed that there are particles of volcanic
origin (clinopyroxene – augite and diopside) in the fabric of
the Porto Recanati amphorae.54 Istrian amphorae rarely con-
tain such volcanic inclusions, and only in trace amounts. The
50
Bezeczky 1997, 178.
51
Mercando 1974, 174; Mercando 1979, 180; Carre 1985, 232-
235; Bezeczky 1987, 35-36.
52
Mazzocchin 2009, 193-195.
53
Bezeczky 2005b, 61; Mazzocchin 2009, 206-207. The stamped
amphorae from Athens and Corinth do not belong to this type.
The Dressel 6A amphorae bear the IVLI (palm) / PAVLINI stamps.
54
Similar volcanic inclusions are found in the Schörgendorfer 558
amphorae too. Though some archeologists thought that the form was
produced on the Istrian peninsula, on the basis of the petrological
analysis, the site of production should be located in northern Italy.
Bezeczky 2005b, 56-57 and 63; Sauer 2005, 120-121 and 123-
124; perhaps in the region of Lake Garda. Pesavento Mattioli
2011 does not agree with this. Their precise origin is unknown but,
based on their distribution, they may have been produced in central
Italy. Research into these amphorae must continue. There are no
Schörgendorfer 558 amphora in the Castrum villa but Vidrih
Perko – Pavletić 2000, 266 suggested it.
BEZECZKY_Text.indd 45
45
Porto Recanati amphorae were clearly produced in north-
ern, or perhaps central, Italy.
What was transported in these amphorae? Olive oil can
probably be ruled out, as it was readily available in Brijuni in
the first and second centuries. Porto Recanati amphorae are
widely known in Cisalpine Gaul and the northern provinces
(Noricum and Pannonia).55
FORLIMPOPOLI / FLAT-BOTTOMED AMPHORA
There are Forlimpopoli / flat-bottomed amphora fragments
from Castrum villa. This means that local wine produc-
tion cannot be ruled out at the villa. It is unclear whether
the wine presses are from the second century or a later peri-
od.56 If there was a significant quantity of wine produced,
as the number of wine presses suggests, then the wine could
have been exported. This would resolve various questions
concerning wine production in Brijuni.57 The petrological
analyses indicate that most of the Forlimpopoli amphorae
came to the Castrum villa from Italy. The analyses of a few
pieces revealed that the Fažana workshop also produced flat-
bottomed amphorae, although the entire form of the vessel
is unknown. Their fabric is similar to that used for the Dres-
sel 6B.58
Workshops have been found in the Forlimpopoli and
Rimini region.59 The most recent discovery was in Crikven-
ica, on the eastern coast of the Adriatic.60 They were used
from the first century A.D. to the end of the third century.61
The content of the amphora was wine. It is widely distributed
in the Adriatic region and throughout the Mediterranean.62
55
Toniolo 1991, 29-31; Mazzocchin 2009, 206-207; Bezeczky
2005b, 61-63.
56
Matijašić 2008, 289-300.
57
Tassaux 1992, 252; Brun 2004, 53-54; Vitasović 2005, 209;
Matijašić 2008, 289-300; Begović – Schrunk 2007a, 42 =
Begović – Schrunk 2007b, 329-330 = Begović – Schrunk
2010, 256 = Begović – Schrunk 2011a, 378.
58
The vessel fragments (handle and bottom) analyzed by Szakmány and
Józsa were found during the excavation of the workshop and in the
Castrum villa.
59
Aldini 1978, 236-245; Aldini 1981, 5-13 and 40-45; Stoppioni
1993; Carre 1985, 228-231; Williams et al. 2005, Forlimpopoli
type; Rizzo 2014, 126-127.
60
Lipovac Vrkljan 2009, 309-314; Lipovac Vrkljan 2011, 9-18.
61
Carre 1985, 230; Williams et al. 2005b, Forlimpopoli type.
62
Carre 1985, 230; Tchernia 1986, 258-260; Williams et al.
2005b, Forlimpopoli type; Rizzo 2014, 127; Bezeczky 2005b, 48;
Auriemma et al. 2008, 175.
18.03.19 14:39
 46
DRESSEL 2-4
Quite a few Dressel 2-4 amphorae were found at the Cas-
trum villa, the majority of which were produced on the
Adriatic coast. Such amphorae were not produced in the
Fažana workshop. Some items were brought from the Tyr-
rhenian coast. These amphorae from Campania (35), have a
very characteristic fabric, with many “black sand” inclusions.
The Dressel 2-4 amphora type was an imitation of the Hel-
lenistic Koan vessels. It was produced at a number of sites
(Italy, Spain, France, North Africa, Cilicia and Egypt), and
its forms are quite varied.63 Places of production have been
found in Rimini/Ravenna and Modena in Emilia Romagna.
Stamps suggest that it was probably also produced in Trieste
and, perhaps, northern Istria (Školarice).64 The production
of the first Dressel 2-4 amphorae is dated to the second quar-
ter of the first century B.C. Production was continuous until
the second or third century A.D.65 This amphora type con-
tained wine.66 They were primarily shipped to the western
Mediterranean, but can be found in the northern provinces
(Raetia, Noricum and Pannonia) and Asia Minor, as well as
at Egyptian, Red Sea and Indian sites.67
SICILIAN AMPHORAE
One fragment of a north Sicilian late Roman amphora (177)
was also found at the villa. Such Crypta Balbi 2 amphorae
are rare in the northern Adriatic region.68 The provenance
of some of the amphorae could not be determined pre-
cisely. Two of them (53 and 54) have the label “Mid Roman
amphorae” in the catalogue. They may come from Sicily too.
63
Panella 2001, 194; Williams et al. 2005a, Dressel 2-4.
64
Žerjal 2008b, 464-465.
65
Williams et al. 2005a, Dressel 2-4; Bezeczky 2013, 129.
66
CIL, IV. 2556-2559, 5554, 5560-5561, 5592, etc.; Panella 1980,
256.
67
Tchernia 1986, 127-129; Panella 2001, 192-211; Williams et
al. 2005a, Dressel 2-4; Tomber 2008, 40; Bezeczky 2013, 129-132;
Rizzo 2014, 108-115.
68
Saguì 1998, 321; see González Cesteros in this volume.
BEZECZKY_Text.indd 46
THE AMPHORAE
CONCLUSION
From the end of the Republican period onward, Istria
and the Brijuni Islands played an ever more significant
role. Wine, olive oil and, perhaps, fish sauces transported
in Adriatic amphorae are among the Castrum villa finds
until the mid-Roman period. The production of the well-
known Istrian olive oil began at the end of the Republi-
can period. Oil production and export was continuous
until the period of Hadrian. A new situation arose when
oil production ceased after Hadrian. We noted that it was
replaced by Baetican oil, and amphorae containing African
oil have also been found. New amphora types (Fažana 1
and 2) appeared, which clearly reflects changes in agricul-
tural production. Wine was initially imported from Italy in
Lamboglia 2, Dressel 6A and Dressel 2-4 amphorae. Later,
wine was also produced on the island, although it is unclear
when production started. The number of presses suggests
that a significant quantity of wine was produced. If this was
exported, one must wonder what vessels were used for this
purpose. As we do not have chronological data, it is impos-
sible to date the Forlimpopoli wine amphorae precisely, and
their origin also requires further study. The Porto Recanati
and Aquincum 78 amphorae, found in limited quantities,
also pose interesting questions.
We do not know much about agricultural production
after the third century, but it is clear that oil production
was continuous. We have no information regarding the
quantities produced, but it is certain that oil was imported
from the eastern Mediterranean in the Late Roman period.
18.03.19 14:39

5 HISPANIC A MPHOR AE
Horacio González Cesteros – Piero Berni Millet
1 IBERIAN AMPHORAE PRODUCTION AND
THE ADRIATIC REGION
The Iberian Peninsula was a very rich region in Antiquity,
famous for the production of huge quantities of basic food-
stuffs and raw materials, mainly metals, grain, horses and
wine, olive oil and fish sauces.1 The last three products were
sent in amphorae throughout the whole Roman world and
beyond its frontiers. Even if we talk in general about a His-
panic production or Hispanic products, the Iberian Penin-
sula was, and is still today, a very varied geographical area,2
with strongly marked contrasts between the north and the
south, between the coastal and inland regions, and between
the Mediterranean Levantine area and the Baetican and
south Lusitanian coastal regions.
These large differences made the Hispanic amphorae pro-
duction widely varied. As containers used primarily for mari-
time or fluvial transport, amphora production was located
in the coastal areas of ancient Hispania, or around some big
rivers such as the Ebro in the northwestern part of Spain, the
Tagus and Sado in the central area of Portugal and, above
all, the Guadalquivir in southern Spain. The three Augustan
provinces of the Iberian Peninsula, Hispania Citerior Tar-
raconensis, Baetica and Lusitania, exported their amphora
products from the Classical period until Late Antiquity,
together with Mauritania Tingitana, a region closely linked
to Baetica since pre-Roman times.
1
The image transmitted by ancient authors from the Roman period,
including Pliny, Strabo and Columela, all show us how versatile the
Iberian economy was.
2
In the third book of his “Geographica”, the Greek writer Strabo makes
a very vivid ethnical and cultural division between the different parts
of the Iberian Peninsula, contrasting a highly developed wine and
oil-consuming Iberia, mainly the south and the Mediterranean coast,
with a barbarian Iberia where the tribes were still living on butter and
acorns.
BEZECZKY_Text.indd 47
47
HISPANIC PRODUCTS IN THE IMPERIAL PERIOD
The strong complexity of the Hispanic foodstuff produc-
tion processes and the variety of the products is manifested
in a clear diversity between the rhythms and markets of the
different Hispanic exports. Concerning the Adriatic region,
so far as we know the Tarraconensis amphorae, mainly for
wine products, are only represented by a probable Dressel
3-2 spike documented in Aquileia3 and another possible
exemplar found in Altino.4 The presence of “Hispanic Dres-
sel 2-4” amphorae is hinted at by some underwater finds
along the Dalmatian coast,5 as for example within the mixed
Spanish and south Italian cargo of the Pupak reef shipwreck
(late first century A.D.), or in the Paržanj island near Hvar.6
In the Pupak reef material, a Dressel 2-4 stamp of HISP has
been interpreted as a sign of a Hispanic production. We have
no parallels for this stamp, and the form of this piece looks
strange for a Tarraconensis Dressel 3-2. Therefore we are
sceptical regarding whether it was produced in any Hispanic
workshop.
Within our material in Brijuni, we found not a single frag-
ment of Tarraconensis amphorae. This total absence suggests
that despite some sporadic pieces found in some contexts
in the central and eastern Mediterranean,7 their main mar-
kets were always located in the western part of the Roman
Empire, above all in southern and western Gaul during the
late Republican and Augustan period,8 and on the Tyrrhe-
nian coast and the cities of Rome and Ostia during the first
century A.D.9
3
Verzár-Bass 1994, 389, Pl. 60 A1.
4
Cipriano 2003, 242.
5
M. Jurišić (2000, 11) remarks that the amphorae from the Školjić-
Unije islands, previously classified as Pascual 1, is not that, but a Por-
torecanati amphora type.
6
Jurišić 2000, 12-14, 70-71.
7
Adamsheck 1979; Hayes 1992; Reynolds 2000; Bezeczky
2008; Bezeczky 2013a, 139; González Cesteros – Yilmaz
2013.
8
Laubenheimer 2015.
9
Tchernia – Zevi 1972; Tchernia 1986; Rizzo 2003; Panella
1973. The problem about finding a good calibration of the quanti-
18.03.19 14:39
 48 THE AMPHORAE
Southern Spain, the Roman imperial province of Baetica,
was an extraordinarily rich area that produced and exported
an impressive agricultural surplus. This success is reflected in
the huge quantities of Baetican amphorae found all around
the Mediterranean, as well as in the central and northern
European regions. The most important Baetican products
exported in amphorae were fish-sauces and salted fish prod-
ucts from the coastal and lacustrine areas, wine or wine sub-
products from all around the province, and vast quantities of
olive oil from the Guadalquivir valley and nearby regions. It
is important to differentiate between the Baetican produc-
tion regions to elucidate the different impacts of Baetican
amphorae imports in the northern Adriatic, and within the
material from Brijuni.
Baetican amphorae production and export are docu-
mented from the Classical period10 to the sixth century
A.D.11 The liveliest export phase of Baetican wares occurred
between early Augustan times and the third century A.D.,
based on the extraordinary quantities of fish-sauces and,
above all, olive oil sent to the military markets in northern
and central Europe, as well as to the civilian subsidized mar-
ket in the city of Rome.
The north Adriatic region shows a specific dynamic,
unlike the Baetican privileged markets in the western Medi-
terranean and closer to its markets in the East. In the north
Adriatic area, we find a good number of Baetican imports
during the first two centuries A.D. Nevertheless, if we inves-
tigate more carefully we can observe varying commercial
trends operating in the different parts of the Adriatic region.
Despite the lack of any complete study for the Dalmatic
area, it looks as if the Baetican products arrived in small
quantities during the first and second centuries A.D. Based
on the underwater and published evidence, we record the
arrival of some Dressel 7-11, Beltrán IIA and IIB, Haltern
70 and Dressel 20.12 Further, from some inland places (but
still close to the coast) such as in Burnum,13 Narona or in
Split,14 we can find Hispanic Dressel 7-11 from the late first
century B.C. or early first century A.D. The Guadalquivir
valley products are still to be found in lower quantities than
ties of Tarraconensis amphorae in the city of Rome was already men-
tioned in a recent paper (González Cesteros 2015). We hope the
most recent works of G. Rizzo, such as the recently published work
about Ostia (Rizzo 2014) can at least partially fill this important
vacuum.
10
Sáez Romero 2010.
11
Bernal 2000; Bernal – García Vargas 2008, 676.
12
Jurišić 2000; Glicksman 2005; Vidrih Perko 2006.
13
Borzić 2011.
14
Glicksman 2005, 1999.
BEZECZKY_Text.indd 48
the fish products in the first century A.D. The presence of
Haltern 70 is only known through four exemplars found in
the Pupak reef finds.15 In the excavations of Diocletian’s Pal-
ace in Split, a fragment of a rim of a first century A.D. Dres-
sel 20 was found,16 but the presence of Dressel 20 is better
confirmed by the discovery of a cargo of second-century
A.D. Dressel 20 pieces in Spinut, near Split,17 many of them
bearing stamps of the Antonine period. They were found
together with some Tripolitana I amphorae,18 clearly indicat-
ing olive oil importation from overseas during the last two-
thirds of the second century A.D. Another Dressel 20 simi-
lar to the Spinut pieces is documented by Cambi as coming
from the island of Žut,19 and a third-century Dressel 20 was
also found on Stipanska Island,20 showing a regular enough
presence of Dressel 20 in this area during the Antonine and
Severan periods.
The presence of Baetican fish-sauce amphorae and some
sporadic second-century Dressel 20 is also documented at
various places along the Istrian Peninsula, as for example in
the waters off Rovijn, where some Dressel 7-11 and maybe
Dressel 20 have been found;21 in Savudrija, where some Dres-
sel 20 fragments are documented;22 in the Loron villa near
Poreč, where Dressel 6B and other Istrian types were dis-
covered, accompanied by some fragments of Dressel 7-11,23
and maybe in Rijeka, where a fragment of a Pompei VII is
mentioned in the Principia excavation.24 Nevertheless, it is
around Pula, the most important city and the Roman port
in Istria, where the biggest quantities of Baetican ampho-
rae within the Istrian Peninsula ought to be found. Indeed,
a recent visit to the Roman amphitheatre in Pula revealed
in the exhibition rooms some exemplars of late second and
third-century A.D. Dressel 20, as well as a couple of exem-
plars of Baetican Dressel 7-11. Most of these amphorae must
come from local excavations in Pula, probably in the har-
bour area. However a late Antonine or Severan Dressel 20,
displayed separately from the other Hispanic pieces, comes
15
Jurišić 2000, 15, 71.
16
Will 1989a, 63.
17
Cambi 1983, 363-381; Jurišić 2000, 22; Radić-Rossi 2007.
18
The non Dressel 20 amphorae from Spinut have normally been clas-
sified as Africana II. Nevertheless, the drawings presented by Cambi
1983 (Figs. 2 and 3) show without any doubt that they are Tripoli-
tanian I forms.
19
Cambi 1983, 382.
20
Cambi 1983, 381-382.
21
Starac 2006.
22
Vidrih Perko – Župančić 2005, Figs. 4, 4-6.
23
Marion – Starac 2001, 120.
24
Višnjić 2009, 130.
18.03.19 14:39
 CHAPTER 5 – Hispanic amphorae (González Cesteros – Berni Millet)
from an underwater find off the coast of Fažana opposite the
Brijuni Islands.25
On the other side of the Adriatic, Baetican amphorae,
mainly for fish products, reached the northern part of Italy
uninterruptedly, constituting a significant portion of the
non-Adriatic amphorae.26 The relatively important quantity
of Dressel 7-11 that arrived here caused, we believe, local imi-
tations to be spawned, developing a special type of Dressel
7-11.27 In the Veneto and Friuli regions proper the arrival of
Hispanic fish sauce-type amphorae are documented only in
small quantities in Aquileia,28 in Altino29 and in the Padova
territory.30 A special case, though, is constituted by the sub-
stantial numbers of these products found in Verona, which
cannot be taken as sporadic finds.31 This pattern of distribu-
tion can be rounded off with the addition of some places in
Emilia-Romagna,32 such as Parma, Piacenza and Modena.33
In most of these last, the pointed Imperial-phase fish-sauce
amphorae (such as Dressel 7-11, Dressel 12 and Beltrán IIA
and B) can be found in good quantities, being reused for the
construction of drainage systems and for the insulation of
floors. On the other hand, the presence of Dressel 20 is not
well documented and it looks like it was a very rare occur-
rence during the first century A.D. We think there are two
reasons that explain this fact. The first one is that in most
of the excavated contexts with Hispanic materials, the His-
panic amphorae were being utilized for some construction
propose, for which the pointed amphorae types were more
useful. Secondly, the high presence of Istrian Dressel 6B
during the first and early second centuries A.D. squeezed all
other kinds of olive oil containers out of the market. A good
25
Cambi 1983, 383.
26
Pesavento Mattioli 2000.
27
This is a question not exempt from controversy. New analyses made by
Capelli excluded the possibility of production of these strange Dressel
7-11 (quite close to Dressel 8) in the central and southern Adriatic,
and probably they are not from a northern Adriatic production either,
even if the last is not to be rejected out of hand. We think the for-
mal features, the presence of ink inscriptions mentioning “Hispanum”,
comparable to the Dressel 7-11 copies from Lyon, and the absence
of these amphorae outside of northern Italy (with the exception of
one exemplar found by Bezeczky in Poetovio) all suggest a produc-
tion centre not too far away from their main distribution places in
northern Italy. For the latest on this subject: Mongardi 2016. For
Poetovio see Bezeczky 1993.
28
Verzár-Bass 1994.
29
Cipriano 2003.
30
Pesavento Mattioli 2000, 744.
31
Pianetti – Modrzewska 1994; González et al. 2000; Pesa-
vento Mattioli 2000.
32
Pesavento Mattioli 2000.
33
M. Mongardi, personal communication.
BEZECZKY_Text.indd 49
49
line of enquiry for the future would be to analyse the pres-
ence of Dressel 20 from the Hadrianic period, once the mas-
sive Istrian olive oil production came to an end and the Baet-
ican Dressel 20 took over the old Istrian markets in Noricum
and Pannonia.34 Aquileia is the best place for achieving a
better understanding of this process.35 Some Dressel 20
have already been found there, some of them with stamps
of the Antonine period,36 and as we have mentioned before,
some shipwrecks with Dressel 20 from the second century
A.D. exist in the Adriatic area.
The westernmost province of ancient Hispania also
exported some amphorae to Brijuni and the Adriatic shores.
The beginnings of amphorae production in Roman Lusita-
nia still remains a contentious subject. Even if an early
amphorae production, imitating the most typical Baeti-
can and Italian forms, was already under way in the second
part of the first century B.C., a wide diffusion of Lusitanian
types is not well documented until the second half of the
first century A.D.37 The main form of this period is with-
out doubt the Dressel 14, produced at a considerable range
of kiln sites on Lusitanian terrain, especially from Peniche
in the north to the River Sado in the south, and along the
Algarve coast region.38 Other forms were developed in the
second and third centuries A.D., above all the Lusitania 3,
a flat-bottomed sort clearly linked to the Gauloise 4 and
Dressel 30 of the second and third centuries.39 The Imperial-
phase Lusitanian amphorae from the first to third centuries
A.D. had a great take-up in their own regional markets, as
well as in some northern Hispanic areas such as Gallaecia40
and in some inland centres of western Baetica.41 Outside the
Hispanic markets, they have been found in Sardinia, on the
route to Rome,42 in Italy in the Campanian area and, above
34
Bezeczky 1998a, 10-11.
35
In September 2013, together with T. Bezeczky, we visited our col-
league F. Schimmer in Aquileia, where he kindly showed us the
ceramic material from his surveys in the canal area, where some Dres-
sel 20 from the second and third centuries A.D. had been recovered.
We are very grateful to Dr. Schimmer for allowing us to see and men-
tion his unpublished material.
36
Pesavento Mattioli 2000, 741. The stamps M.E.EVPRO and
M.AEM.RVS have been classified as of Antonine production by Berni
(2008, no. 1367 and 1279).
37
On the diffusion of Lusitanian Amphorae, see in the recent publica-
tion: Vaz Pinto et al. 2016.
38
Fabião 2008, 729-733; Vaz Pinto et al. 2016.
39
Diogo 1987; Fabião 2008; Quaresma – Cordero Raposo
2016; García Vargas 2016.
40
Fernández Fernández 2016.
41
García Vargas 2016.
42
Porqueddu et al. forthcoming.
18.03.19 14:39
 50 THE AMPHORAE
all, in Rome43 and its port in Ostia.44 They sporadically reach
some places in the eastern Mediterranean, for example Bei-
rut45 or Elaussia Sebaste,46 but in the Adriatic there were no
real regular deliveries during the Roman Imperial period.47
HISPANIC PRODUCTS IN LATE ROMAN TIMES
The export economy of ancient Hispania was transformed
during the third century A.D., undergoing several changes
that not only affected the main amphora forms, but also
their markets and routes.48 During the Late Empire, the Bae-
tican products still appear in large quantities in their primary
markets in the western part of the Empire,49 but not at the
overwhelming level they did before. In the eastern Mediter-
ranean, even if the preliminary works of D. Bernal50 have
laid a foundation for further research, we still lack detailed
studies about the Hispanic amphorae diffusion in the dif-
ferent eastern regions. The situation is best known in places
such as Corinth,51 Ephesus,52 Gortyn,53 Beirut,54 Caesarea
Maritima55 and Alexandria,56 which reveal a continuous, but
minor, trickle of Baetican products from the Augustan era to
the first half of the fifth century A.D. and even later, as the
presence of some Keay 19 and Dressel 23d in contexts with
material predominantly dating from the late fifth and sixth
centuries A.D. in Ephesus have recently suggested.57 In the
case of Lusitania, the production of their amphorae reached
its peak in the fourth and fifth centuries A.D.58 and contin-
ued, at a diminished level during the sixth century too.59
For the northern Adriatic area, we can confirm a continu-
ous presence of Hispanic amphorae (at this moment exclu-
43
Rizzo 2003; Martin 2016.
44
Panella 1973; Rizzo 2014; Martin 2016.
45
Reynolds 2000.
46
Ferrazzoli 2010.
47
The presence of Dressel 14 is documented by some sherds at Aquileia:
Gaddi – Degrassi 2016, 440, but it was not possible to determine
if they were of Baetican or Lusitanian origin.
48
Bernal Casasola 2000a; Bernal Casasola 2000b.
49
González Cesteros 2010; Berni Millet – Moros Díaz 2012,
214-217.
50
Bernal Casasola 2000b; Oren-Pascal – Bernal 2000.
51
Warner Slane – Sanders 2005; Will 1983.
52
Bezeczky 2008; Bezeczky 2013a; González Cesteros 2012;
González Cesteros – Yilmaz forthcoming.
53
Romero – Pascale 2001.
54
Reynolds 2000.
55
Oren-Pascal – Bernal 2000.
56
Will 1983.
57
This material is still being processed by one of us; it is hoped that we
will soon be able to present it to the scientific community.
58
Fabião 2008, 740-743.
59
Fabião 2008, 743.
BEZECZKY_Text.indd 50
sively Lusitanian and Baetican products) in contexts from
the third and fourth centuries A.D.60 They disappear from
the Adriatic progressively during the first half of the fifth
century A.D.61
Olive oil amphorae from Baetica are very rare in our
region during the late Roman period. We have only found
references to Dressel 23 in northern Istria, as, for example, in
the villa of Školarice near Koper in Slovenia,62 and maybe in
Savudrija, in the coastal area of Croatia.63 In the Trieste and
Venetia environs, we also find some references to a sporadic
presence of Dressel 23.64 Forms related to fish products,
such as Keay 16 and Keay 19, are present almost everywhere
in the Adriatic, but always in small quantities.65 An example
of the level of the influx of these kinds of amphorae in the
Istrian region is the discovery of a complete Almagro 51C,
probably of Lusitanian origin, in waters near Barbariga,
some kilometres north of Fažana,66 and the recent discover-
ies in the harbour area of Pula.67 Together with these pieces
in Aquileia are some Beltrán 72,68 and here again, and in
Trieste and in our context in Brijuni, also some Beltrán 68.
Both amphora types are interesting Baetican ones linked to
fish products and wine exportation, from the third to early
fifth centuries A.D.69
Regarding the Lusitanian amphorae, it remains true that
an accurate differentiation amongst the various Baetican
products is seldom met with in publications on the Adriatic
and other Mediterranean regions, as has been mentioned sev-
eral times and recently by virtually every scholar present in
the workshop that took place in Troy in October 2013.70 In
60
Auriemma et al. 2012.
61
The practical absence of Hispanic amphorae in Classe (port of
Ravenna) in contexts from the fifth century A.D. is quite remarkable.
In the southern Adriatic, some small quantities of Almagro 50 and
Keay 19 have been found in contexts from the fourth and fifth centu-
ries (Volpe et al. 2010), but later they disappear completely from the
Adriatic region.
62
Žerjal 2010, 704.
63
Vidrih Perko – Župančić 2005, Figs. 4, 5.
64
Maselli Scotti et al. 2003, 25; Auriemma et al. 2012.
65
Jurišić 2000, 22; Vidrih Perko 2006; Auriemma et al. 2012.
66
Višnjić et al. 2010.
67
We would like to thank the excavation director Ms. Ida Koncani Uhač
for inviting us to see this interesting material from the second to early
fourth centuries A.D.
68
Gaddi – Degrassi 2016, Fig. 5.
69
For Beltrán 72: Bernal 2016b; González Cesteros et al. 2016;
for the Beltrán 68: Bernal Casasola 1996; Bernal 2009; Ber-
nal 2016a. In Trieste they found a complete Beltrán 68 in the exca-
vations of a well in the Roman forum, but it was identified, together
with a complete Baetican and some Lusitanian Almagro 51C, as a
Mauritanian Dressel 30 (Degrassi – Maggi 2011, Figs. 1, 3-5).
70
Vaz Pinto et al. 2016.
18.03.19 14:39
 CHAPTER 5 – Hispanic amphorae (González Cesteros – Berni Millet)
most cases, scholars have published them en bloc, assigning
the whole Hispanic production, with the exception of olive
oil containers, to the Lusitanian. This problematic question
will be tackled next. For now, we can only point out that,
due to the common use of the same maritime routes,71 we
should expect the presence of types like Almagro 50 (Keay
16, Keay 22), Almagro 51A-B (Keay 19), and Almagro 51C,
that is, a dual presence of Baetican and Lusitanian pieces at
the end markets, as well as occasionally finds of other Lusita-
nian forms, as, for example, the Keay 78 found in the Forum
of Aquileia.72
2 IDENTIFICATION PROBLEMS IN DISTIN-
GUISHING LUSITANIAN AND BAETICAN
AMPHORA TYPES
The close connections between the Baetican and Lusitanian
amphora types have their roots in the facts that both prov-
inces are located on the Atlantic seaboard and in the eco-
nomic expansion of Baetican production from pre-Roman
times. A close constant economic relationship existed
between both provinces, and the production of similar
amphora types in both regions runs from the second half
of the first century B.C.73 down to the latest documented
exports in the sixth century A.D., or even later.74
The first Lusitanian Roman amphora types followed the
manner of the ovoid amphorae, close to those produced in
the Guadalquivir and Cadiz workshops during the second
half of the first century B.C.75 Accordingly a Lusitanian ver-
sion of types such as Haltern 70 and Dressel 7-11 in Augus-
tan times has been proposed in several publications,76 and a
limited production has been documented in the territory of
Augusta Emerita.77 Nevertheless, before the Julio-Claudian
dynasty the Lusitanian amphora exports were mostly limited
to the provincial and nearby Baetican and Galician markets.
It is during the second part of the first century A.D. that the
containers from the Sado-Tagus region and the Algarve (this
last region was even more closely linked to Baetica) start
to be produced in huge quantities and reached customers
beyond the peninsular borders.
71
Bombico 2016; Bernal 2016c.
72
Vide infra.
73
Morais – Fabião 2007.
74
Bernal – García Vargas 2008, 676.
75
Morais – Fabião 2007; García Vargas et al. 2011.
76
Fabião 2008, 726.
77
Bustamante Álvarez – Heras Mora 2013.
BEZECZKY_Text.indd 51
51
The Lusitanian amphora production was dominated dur-
ing the last part of the first century and the second century
A.D. by the Dressel 14. This amphora is the most typical
Imperial type produced in Lusitania, but at the same time
a Baetican variant exists too, with some typological pecu-
liarities. In the 1970s scholars such as M. Beltrán78 and
A. J. Parker79 discriminated between them, calling them A
and B. Parker named Dressel 14B as the Lusitanian version:
still today this remains the most commonly used name for
this variant. Even if scholars progressively observe the dis-
tinction between the two variants, there yet remains an
unremitting confusion. This lack of clarity gets in the way
of a better understanding of the productive and commercial
dynamics of both provinces during the first and second cen-
turies A.D. We think that the Baetican version was produced
at a lower level than the Lusitanian Dressel 14, being a sec-
ondary type lagging behind other main fish-sauce vase-types
from Baetica in the late first and second centuries A.D., such
as Beltrán IIA and IIB.
For the most important Lusitanian and Baetican amphora
types of Late Antiquity, the same problems of duality and
resemblance are even worse than during the Imperial period.
Types such as Almagro 50, 51A-B and 51C were produced
in a large number of kilns from today’s Galicia to eastern Bae-
tica, resulting in major formal variations, as already noticed
by Keay in his work about Late Antique amphorae in north-
eastern Spain.80 However, since the end of the 1990s the
development and study of the very rich amphora production
of Late Antique Hispania has clarified the typological differ-
ences and the individual commercial dynamics of Lusitanian
and Baetican production. The work demonstrates not only
the formal variants linked with each one of these regions,
but also the pre-eminence of certain types among each pro-
duction family. The Lusitanian potters might choose the
Almagro 51C and some particular versions of the Almagro
50, such as the Keay 22 and the so-called “Sado 1”, as their
principal and preferred types,81 while in the Baetican work-
shops the Keay 19 and the bigger Keay 16 were favoured.
These generalities do indeed assist in giving a better distinc-
tion between both producers on a regional basis, but the
reality is more complex still. We must consider the implica-
tions about the exceptional numbers of workshops involved
in the production of amphorae, each producing several dif-
78
Beltrán 1970, 456-464.
79
Parker 1977, 37-39.
80
Keay 1984, with all the variants proposed for forms 16, 19, 21, 22
and 23.
81
Fabião 2008, 736-743.
18.03.19 14:39
 52 THE AMPHORAE
ferent amphora types at different times. A better knowledge
of the fabrics and an improved familiarity with the Hispanic
forms is the only safe way to establish a certain identification
of Baetican and Lusitanian types.
Another important typological misunderstanding affects
some special Hispanic types retrieved at different Mediter-
ranean sites: those classified as Mauritanian Dressel 30. This
problem particularly impacts upon the Baetican Beltrán 68,
a type produced in various Baetican workshops from the
third to the early fifth centuries A.D.82 It has turned up here
and there in Italy and the eastern Mediterranean, for exam-
ple in Caesarea Maritima,83 Gortyn,84 Brescia85 and Aquileia,
where at least three complete examples have been found.86
These amphorae were classified as Dressel 30 in Gortyn and
Aquileia, as well as in a recent publication of the ceramic
material from Caesarea,87 even though Bernal had already
published them as Beltrán 68.
In our opinion, the Hispanic amphorae of the Forum
context in Aquileia have been badly misjudged. Not only
was the Beltrán 68 published as a Mauritanian Dressel 30,
but so also were a complete Baetican Almagro 51c/Keay
23,88 as well as some third/fourth century Almagro 51c/
Keay 23 (with a large everted rim and flat handles with cen-
tral groove).89 Furthermore, in the old Forum excavation a
Lusitanian Keay 78 was classified as a Dressel 23.90
It is necessary to make a careful revision of Hispanic
amphora imports from the late Julio-Claudian period right
through to Late Antiquity. Paying attention to their for-
mal features and to the macroscopic characteristics of their
fabrics constitutes the only way to formulate a better dis-
tinction between the Baetican and Lusitanian amphorae in
their eastern and central Mediterranean places of import.
We fully acknowledge that it is not an easy matter, because
there are a lot of resemblances between types and subtypes
and a perplexing range of Baetican and Lusitanian fabrics,
but today good works of synthesis and internet tools exist
82
Bernal 2016a.
83
Bernal Casasola 2000b.
84
Portale – Romeo 2001.
85
Bruno 2002, no. 63-64.
86
Degrassi – Maggi 2011, Figs. 1, 3; Villa 2013.
87
Johnson 2008.
88
This piece should have been produced in the same Cadiz work-
shops as the Beltrán 68, as resemblances in shape and fabric declare
(Degrassi – Maggi 2011, 263, Figs. 1. 3 and 4).
89
This kind of Almagro 51c has recently been found in some new exca-
vations in the harbour area at Pula.
90
Verzár-Bass 1994, Pl. 60, AS 6.
BEZECZKY_Text.indd 52
by means of which one can at least achieve a correct typo-
logical identification.91
3 THE HISPANIC AMPHORAE FROM THE
BRIJUNI CASTRUM VILLA
After presenting a general overview of the Adriatic region
and some problems related to the Hispanic production, it
is time to analyse the Hispanic material documented during
our campaigns on the island of Brijuni.
Within the material from the Castrum where we worked
on Brijuni, we retrieved a small quantity of Hispanic ampho-
rae. Even if the bulk of local and Adriatic products outnum-
bers the finds of imports from other regions, the Hispanic
material is still found in much lower quantities than the Afri-
can or the Oriental containers. As was the case for the whole
Istrian and eastern Adriatic region, the small number or lack
of documented Hispanic amphorae till late Julio-Claudian
or Flavian times can be partially explained by the existence
of locally produced olive oil and fish products.
Nevertheless, for olive oil, we find within the Castrum
material some Baetican Dressel 20 (40-45). All of them are
variants produced after the time of Hadrian’s reign, and are
of the typical Guadalquivir Valley fabric. From the second
century A.D., together with the almost complete amphora
(41) with a graffito ante cocturam presented by P. Berni in
this volume, we could recognize some other rim and handle
fragments of Dressel 20, one of them bearing a stamp from
the second half of this century.92 During the Severan period
and the third century A.D., the Dressel 20 continued to
arrive in Brijuni, but for the fourth and fifth centuries we
lack any Dressel 23 or nearby Baetican olive oil types. This
pattern seems to be repeated throughout the whole Adri-
atic region, where Dressel 23 are really rare. The explanation
must be sought in the Adriatic’s own production and in the
new sources of olive oil imports, such as, for example, Africa
and the eastern Mediterranean.93
91
A very useful tool for any amphora specialist is the internet-site of the
project “Amphorae ex Hispania. Paisajes de producción y consumo”:
http://amphorae.icac.cat/. Here the archaeologist can find a detailed
description with wide-ranging literature about the most important
amphora types produced in the Iberian Peninsula and the Balearic
Islands.
92
Catalogue no. 40, see Amphora stamps and inscriptions chapter.
93
Some oriental amphora types possibly linked to an olive oil content,
such as the LRA 2, arrived in the Adriatic area, but their quantities
are not really enough to represent a continuous supply of oil from this
region. The African amphorae are well-documented from the third
century A.D., but not all of the present types have to be linked to an
olive oil content (see Bonifay – Capelli in this volume).
18.03.19 14:39
 CHAPTER 5 – Hispanic amphorae (González Cesteros – Berni Millet)
We do have some earlier Hispanic fish-sauce amphora
exemplars in Brijuni. From the late first century A.D. there
is a rim fragment of a late form of Dressel 7-11 (37), similar
to some pieces found in the Neronian-early Flavian depot of
“Estagnon” in Fos-sur-Mer.94 The fabric of our piece is buff
in colour on the external surface, also with traces of a white-
cream slip, and red brick colour at the break, with a hard and
sandy texture: it is close to the red-coloured products of the
Cadiz area. The Beltrán IIA base (38) must be dated from the
same period or a little bit later. The fabric of this piece is buff
in colour on the surface and at the break too, of a relatively
soft texture with many small quartz inclusions, small voids,
and some hematite inclusions. These petrological features
suggest a different but nearby production area, maybe on the
banks of the ancient Lacus Ligustinus or in the area around
today’s Jerez de la Frontera. Also from the late first century
or the first half of the second century A.D. is a rim that seems
to be from a rough Dressel 14A (39).95 This type was pro-
duced mainly in the Cadiz-Algeciras and Malaga-Granada
areas from the Julio-Claudian to the late Antonine periods.96
The fabric of our exemplar seems to be closer to some pro-
ductions of the Malaga region: it is hard and coarse, buff in
colour on the surface and at the break, with large quantities
of small and medium-rounded quartz of different sizes, some
sparse small red pieces and large lime-based inclusions; some
big voids are seen at the break. The typological features seem
close to the Malacitanian Dressel 14, such as were produced
in the Calle Carretería workshop.97 We can argue that the
zenith of the production of Dressel 14 in the Mediterranean
Baetica is around the Flavian period and the first half of the
second century A.D. This was probably the moment when
this piece arrived at Brijuni.
Wine from Baetica is also documented in Brijuni by the
presence of two exemplars of Beltrán 68, one of them (51)
undoubtedly produced in a workshop in the Bay of Cadiz in
the late third or fourth century A.D. The fabric of this exem-
plar shows, without a doubt, its origin in the Bay of Cadiz,
and is similar to the late Dressel 7-11 rim (37). It is still cov-
ered by a pale slip on the external surface, but the colour is
red brick at the break and on the inside. The fabric is hard
and coarse and has several large hematite inclusions.
94
Marty – Zaaraoui 2009, Figs. 9,3; 10,3.
95
We would like to thank Dr. García Vargas for helping us with this
form.
96
Bernal Casasola 1998; Mora Serrano – Corrales Aguilar
1997; García Vargas 2000, 83-84.
97
Rambla Torralvo – Mayorga Mayorga 1997, Figs. 4-5.
BEZECZKY_Text.indd 53
53
For the late Roman period we have some Hispanic fish
amphorae in the material from the Castrum. Despite the
absence of Lusitanian products during the first two centuries
A.D. and the continuous arrival of Baetican amphorae then,
during the fourth and probably fifth centuries the Lusita-
nian products replaced the Baetican imports being sent to
Brijuni, and all of the documented fish amphorae from this
period come from Lusitania. The presence of rims from two
versions of the Lusitanian Almagro 51A-B (47 and 48) must
be placed at the end of the fourth century A.D. or, better
still, during the fifth century A.D.,98 a time of high produc-
tion and export from the Lusitanian workshops. The fabric
of these two exemplars is different in colour and texture, but
in both cases they are of a central Lusitanian production.
The first Almagro 51A-B rim (47) exhibits a red colour with
remains of a white slip; it is hard and coarse, with many small
and medium-size rounded or sub-angular inclusions, most
of them quartz of different colours; it also has occasional
small mica flocks, some lime inclusions and a couple of red
rounded inclusions, probably grog/chamotte. On the other
hand, the other Almagro 51A-B, with a longer rim, is in a
finer fabric, buff in colour, soft and sandy in texture and with
a large number of small lime and quartz inclusions. It also
has some occasional mica flocks and rounded and small red
inclusions. This fabric is similar to the rim of the Almagro
51C (49), just the colour is a little bit different. The other
Lusitanian amphora we have documented in Brijuni is a ring
base of form Keay 22, a related type to the Almagro 50, pro-
duced in the Lusitanian workshops during the fourth and
fifth centuries A.D.99 In this case the fabric is also close to the
last two fragments, but it was fired at a higher temperature
that ensured the disappearance of almost all the lime compo-
nents, creating some large voids. The external surface is pale
buff in colour and at the break it is buff and orange, present-
ing the small and medium-size red inclusions and occasional
mica flecks typical of Lusitanian production.
CONCLUSION
In conclusion, we can argue that the Hispanic amphorae
arrived at Brijuni from the late first to the early fifth centu-
ries A.D. Even if they travelled together, the dynamics of the
Baetican and Lusitanian imports suggest that the import of
the products of these two Hispanic regions occurred at dif-
ferent times. The Baetican amphorae arrived mainly in the
98
Mayet – Tavares Da Silva 1998, 204-205; Fabião 2008, 740.
99
Fabião 2008.
18.03.19 14:39
 54 THE AMPHORAE
second and early third centuries A.D. This is the case of the
olive oil containers that always appear after Hadrian’s prin-
cipate, at the abrupt end, then, of the Istrian olive oil pro-
duction.100 For the fish products it is important to stress
the absence of early Dressel 7-11 forms. We know they
arrived all the while in the whole Adriatic region, and even
if they are not very numerous in the Istrian Peninsula, some
examples of first century A.D. Dressel 7-11 do exist in Pula,
Rovijn and Loron. For Brijuni, the earliest fish amphorae
type is the late Dressel 7-11 rim from the Neronian or early
Flavian period. Equally, one should underline the absence of
late Roman Baetican fish amphorae types, such as Keay 16
and Keay 19, which is somewhat surprising as they are docu-
mented in other Adriatic contexts.101 The varied Baetican
forms we have in Brijuni reflect the different places where
these amphorae were produced inside the Baetican territory.
All of the olive oil containers have the typical Guadalqui-
vir valley fabrics, while the fish-amphora types show a mul-
tiplicity of fabrics from Malaga to Lacus Ligustinus, with a
central focus around the Bay of Cadiz.
In the case of the Lusitanian products, they arrived from
the fourth century A.D. and continued at least during the
first half of the fifth century. All of them are fish amphorae
types: they appear to replace the Baetican imports of the
100
Bezeczky 1998a, 10-11.
101
Jurišić 2000, 22, Pl. 4; Auriemma et al. 2012.
BEZECZKY_Text.indd 54
previous centuries. Despite the different types documented
(Keay 22, Almagro 51C and Almagro 51A-B), all of them
seem to come from the same Lusitanian region: this is situ-
ated in central Lusitania, from the workshops around the
Tagus and Sado confluences.
Even if the Hispanic material arrived only in minor quan-
tities at Brijuni, it nonetheless formed a continuous presence
from the late first century A.D. to the fifth century A.D. This
is more or less the same image we have gleaned from other
Dalmatian and north Adriatic contexts, but the absence of
some Baetican fish amphorae types of the Augustan and
Julio-Claudian periods, as well as late Roman Baetican fish
amphorae, is remarkable. On the other hand, the presence of
Dressel 20 from the second and third centuries is symptom-
atic of the decline of the Istrian olive oil production from
the late Hadrianic period: this pattern supports the theory
of a reduced production of Istrian oil amphorae and the
arrival of Baetican products in the north Adriatic and in the
Danubian military markets.102 In the case of fish products,
the local production must have been constant during the
entire Roman period and beyond, but Hispanic products,
first from Baetica and later from Lusitania, were imported
into this particular place in the northeastern Adriatic at least
until the fifth century A.D.
102
Bezeczky 1998a, 10-11; Bezeczky 2000.
18.03.19 14:39

6 E ASTERN MEDITERR ANE AN A MPHOR AE
Horacio González Cesteros
THE EASTERN MEDITERRANEAN AMPHORAE
AND THE NORTH ADRIATIC REGION
Archaeological finds all along the Adriatic region show the
existence of a special link between this geographical zone and
the eastern Mediterranean during Antiquity. The proximity
to western Greece and the Aegean world through the Ionian
Sea and then the Corinthian Gulf and Isthmus supported
a continuous commercial activity, in both directions, docu-
mented well back into pre-Roman times.1 Not only ampho-
rae imports were involved, but many other types of pottery
and diverse goods, for example marble and other kinds of
stone.2 Nevertheless, the commercial trends that linked the
Adriatic region and the eastern Mediterranean seem to have
taken differing paths between the north and south Adri-
atic, mainly during the late Republican and early Imperial
periods. We can interpret this distinction as part of a trade
directly from both regions into the Aegean area, something
also easily observed through the different amphora types
produced in each zone in Roman times. For example, in
Apulia the main amphora factory, concentrated in the sec-
ond and first centuries B.C., produced its own ovoid forms
with close relations to the north Peloponnese amphorae that
developed from the so-called Corinthian amphorae, pro-
duced in Corinth and Corfu from the Archaic to the Hel-
lenistic periods.
It is a commonplace observation that in the western Med-
iterranean research does not distinguish enough between
the materials emanating from the different eastern Mediter-
ranean regions. The traditional western viewpoint assigns
the classification of all the eastern material within the broad
term “eastern Mediterranean amphorae”. Perhaps this atti-
tude can be understood for places such as Gaul, northwestern
1
There exists a huge bibliography about Greek imports in the
region before the Roman conquest: AAD 1977; Brusić 1999;
Auriemma 2006; Auriemma 2007; Slane 2008; Pizzirani 2013;
Auriemma – Silvestrilli 2013; Ugarković 2016; etc.
2
Jurišić 2000, 39-41; Carre – Maselli Scotti 2001, 230;
Glicksman 2005.
BEZECZKY_Text.indd 55
55
and central Europe or even the Iberian Peninsula, but not for
the Adriatic region, where imports reach high percentages
from the Hellenistic period, or even earlier. We believe it is
somewhat important to refine the classifications, because we
are dealing with a vast geographical expanse, with a huge cul-
tural, political and geographical diversity.3
1 REPUBLICAN PERIOD
During the late Republican period the commercial relations
between the Adriatic and the Aegean increased in a signif-
icant way. This was due to several factors, for example the
widespread commerce linked with the Roman occupation
of the northern Mediterranean territories throughout the
second century B.C., or the establishment of a free port at
Delos in 167 B.C. – a move that was a serious inducement for
exchanges between Italy and the Aegean. So far as concerns
the Adriatic region, an important factor for the economic
and commercial growth was the colonization of broad areas
all along the coast and the most important valleys, for exam-
ple the Po, as well as the control of Illyrian piracy in the late
third and second centuries B.C. Archaeological work has
brought to light the introduction of big quantities of some
amphora types from the Adriatic to the Aegean during the
Republican period,4 but the arrival of Aegean wares in the
Adriatic area seems even more important, above all from the
moment the new Roman province of Cisalpine Gaul started
to be colonized and developed.5 This involves the late Hel-
lenistic Greek amphorae found all along the Adriatic region,
3
A new study about the Cilician, Cypriot and Levantine pottery
imports from latest excavations of Padova University in Aquileia has
recently been published (Dobreva – Riccato 2015). We think
it a good idea to focus more on the different eastern Mediterranean
regions, even more so for sites producing such large amounts of mate-
rial as Aquileia.
4
Grace – Savatianou-Pétropoulakou 1970; Tchernia 1986,
68-74; Lund 2000; Bezeczky 2006; Bezeczky 2008; Bezeczky
2010.
5
Baldaci 1972.
18.03.19 14:39
 56 THE AMPHORAE
mainly Rhodian,6 but also others such as Koan, Knidian or
Chian.
2 EARLY AND MIDDLE ROMAN PERIOD
During the Imperial period we can observe the presence of
a high number of eastern amphorae, often the most signifi-
cant amphora imports after the Italian,7 with a richer variety
of forms and a bigger range of provenances. For the Istrian
Peninsula they do turn up in several places, such as, for
example, Pula,8 in the Roman villa of Školarice,9 Loron,10 or
in Rovijn,11 but as yet the systematic and quantitative analy-
sis of most Istrian contexts is inadequate. We are trying to
fill this gap now, and hope that the most recent discoveries
in the harbour area in Pula will bring more precision to the
matter.
Some amphora types from areas outside the Aegean, such
as, for example, Palestine or Cilicia, have also been found all
along the Adriatic, the northern area being no exception.12
This is a trend that developed still more from the second
century A.D., reaching its highest point in Late Antiquity
with the arrival of large quantities of Aegean, Levantine and
Cilician vessels and the sporadic presence of products from
other regions such as Crete, the Black Sea or Egypt. Never-
theless, from the time of Augustus to the mid-second cen-
tury A.D., the bulk of the oriental amphorae in the north
Adriatic area arrived from the Aegean or nearby regions.
We also find new amphorae production regions repre-
sented in the northern Adriatic from early Imperial times.
Even if amphorae from places such as Knidos, Kos, Chios
and above all Rhodes13 (or nearby territories) continued to
be imported into the Adriatic in good quantities, comprising
the bulk of the eastern Mediterranean imports, other pro-
6
Jurišić 2000, 14; Tiussi 2007; Marengo – Paci 2008. It has been
suggested that local imitations of Rhodian amphorae were produced
in the late Republican period in northern Italy (Baldacci 1972, 125;
Facchini 1997), based on the interpretation of single petrographi-
cal analyses from Calvatone (Facchini 1997) and other places. We
prefer to exercise caution on this controversial subject.
7
Jurišić 2000; Pesavento Mattioli 1992; Pesavento
Mattioli 2005; Cipriano 2003; Glicksman 2005; Cipriano –
Mazzocchin 2011; Mazzocchin 2013.
8
Personal visit to the museum.
9
Žerjal 2008.
10
Marion – Starac 2001, 119-120.
11
Maric 2009.
12
For new finds in Aquileia: Dovreba – Riccato 2015, 132.
13
A new study about Imperial Rhodian amphorae on the Croatian coast
is being carried out by P. Dugonjic, who presented new and interest-
ing material from the early Imperial period, some of it from underwa-
ter finds: Dugonjic 2015.
BEZECZKY_Text.indd 56
duction centres such as Crete or the Asia Minor coastal area
were also exporting their products to our region. Amphorae
such as the Cretan AC4, AC1 and in lesser quantities AC3
and AC2, as well as the Asia Minor versions of the “one-
handled jar”, including early Agora F 65/66 versions,14 are
more or less normal on both sides of the Adriatic in this
period. Other forms, for example the first and early second
century Dressel 24 probably produced in the Chios/Eryt-
rai region, have also been found, albeit but sporadically,15
as well as the alum amphorae from Milos.16 This last form
is a specific import answering the needs of the local wool
industry.17
A more problematic matter involves the small, but con-
tinuous presence of form Dressel 25 in some Adriatic places
such as Trieste18 and, above all, in some Cisalpine cities.19
This type is present in the Dressel table made up mainly from
the Castra Praetoria amphorae, a corpus which is primar-
ily dated to the mid-first century A.D. Its production has
normally been linked with the late Corinthian and ovoid
Apulian production due to its typological characteristics:
this has recently been confirmed by finds in the northern
Peloponnese, among them the pottery workshop of Aigion,
where related forms were produced from the second until
the late first centuries B.C.20 The epigraphic evidence found
on these amphorae, both tituli picti and stamps, always
appears in Greek. The fabric analysis carried out on some
exemplars in Padua21 and in Ephesus22 determines their pro-
duction on the Greek mainland, in the Corinthian hinter-
land or the northern Peloponnese. This new archaeological
data allows us to definitely rule out Manacorda’s suggestion
14
Berti 1990; Carre 2007, 591.
15
Radić – Jurišić 1993; Jurišić 2000; Belotti 2008b.
16
Cipriano et al. 2005.
17
Hence the relatively large number of alum amphorae from Milos
found in Padua: all linked with the important wool industry of this
city (Pesavento Mattioli 2005). They turn up in other places in
northern Italy, for example Novara in the Upper Po valley (Quiri –
Spagnolo Garzoli 2015).
18
Auriemma 2007, 141.
19
Cipriano et al. 1997; Cipriano – Ferrarini 2001, 76-78;
Mazzocchin 2013, 87-88. This form is above all well-represented in
Vicenza (Mazzochin 2013, 87-88), already making its way into the
interior of the Po valley. Starac and Matijašić have also pointed out the
presence of one rim of a Dressel 25 in a depot, mainly stocked with
Lamboglia 2 and Dressel 6B, in Pula (Starac - Matijašić 1991).
However, we are not sure this fragment can really be interpreted as a
Dressel 25 or Class 24 of Peacock and Williams: it looks much more
like a Brindisian form.
20
We would like to thank K. Filis for this important information and
the permission to see this important material in Aigion. Filis 2016.
21
Cipriano et al. 1997.
22
Bezeczky 2013a, 91-93.
18.03.19 14:39
 CHAPTER 6 – Eastern Mediterranean amphorae (González Cesteros)
about a production place around ancient Venafrum,23 based
on the elevated numbers of this form present in Ostia.
From the late second to the early fourth centuries A.D.,
good quantities of a varied range of eastern Mediterranean
amphorae continued to arrive in the northern Adriatic. This
chronological time-frame reflects an interesting period in
the ceramic production all around the Roman world: it is
marked by the diffusion of some amphorae types developed
from old prototypes, but also by the creation of new forms
that are going to form the basis for the typical productions
of Late Antiquity. We find some good dated contexts of this
period in the northern Adriatic, mainly in Trieste24 and in
the Aquileia region,25 but also in Pula, where a high num-
ber of the amphorae from the museum, today exhibited in
the Roman amphitheatre, certainly belong to the third and
early fourth centuries A.D. The material from the excava-
tions near the harbour that we were able to see in the archae-
ological museum in May 2014 is also predominately to be
dated to the late second and third centuries.26 During this
important phase the region must have been under the direct
influence of the new and significant role played by Spalato
and Aquileia as important centres within the organisation of
the Roman Empire. As for commerce with other regions, a
significant increase in the African imports is to be detected,
mainly terra sigillata and several types of amphorae: they
appear in quantities equal to or even in excess of the oriental
vessels. Concerning local amphorae production, from the
second half of the third century that of the Adriatic drops
to its lowest level: the region becomes definitely marginal in
amphora production and export.
The eastern Mediterranean forms continue to arrive in
good quantities, above all the Aegean products. Neverthe-
less, the most important oriental imports of the Hellenis-
tic and early Roman phases, namely the Rhodian and Koan
amphorae, gradually disappear from the mid-second cen-
tury A.D. onwards, due to a general economic trend that
affected the whole Roman world. The vacuum left by this
event is filled by material from a variety of other geographi-
cal zones, mainly the Asia Minor coastal area and probably
some nearby islands like Samos and Chios, where the dif-
ferent versions of the “one-handled jar” (Athenian Agora
23
Manacorda 2012, 153.
24
Auriemma – Quiri 2004; Auriemma 2007; Auriemma et al.
2012.
25
Degrassi – Maggi 2011; Auriemma et al. 2012; Villa 2013;
Dovreba – Riccato 2015.
26
We would like to thank Ms. Ida Koncani Uhač for allowing us to see
this material.
BEZECZKY_Text.indd 57
57
M 45; M 125-126; J 46; J 47; M 240; M 255)27 and other
amphora types, such as the early versions of M 273 and
above all the late Dressel 24 or related types, were originally
manufactured.
The Mediterranean coast of Anatolia also started to
export amphora products regularly to the whole Roman
world: even if they only arrived in a roundabout way in
our region, they have nonetheless been found in almost
every archaeological context. From the southern Anatolian
regions comes a series of amphora forms with a common
characteristic, namely a bifid or pseudo-bifid handle. Even
if one admits to an Aegean production of the so-called
Pseudo-Koan and pinched bifid types during the second
century A.D., it is undeniable that they were also produced
in regions such as Pamphilia, Isauria and Cilicia. The same
can be said for the different variants of Agora G 199 and
M 239, one of the most frequently encountered amphora
types of the eastern Mediterranean from the late first to the
early fourth centuries: they were probably also produced in
Asia Minor28 and a Cypriot production has also been moot-
ed.29 On the other hand, Agora M 54 and G 198 are typical
forms from Isauria, Cilicia and Cyprus30 that also reached
our region sporadically.
Amphorae from other eastern territories only arrived in
really insignificant numbers in the northern Adriatic during
this period. Pieces coming from the Black Sea, mainly from
Sinope, have been retrieved,31 but always in inconsequential
amounts, likewise the Levantine amphorae. Even if their
presence is important to show the arrival of these goods, we
must admit that they were far from forming any part of the
bulk of amphora imports into the most important places in
northern Dalmatia, Istria and Cisalpina. On the other hand,
some Levantine products had already arrived even in the first
century A.D., thereafter being documented during the early
and middle Imperial Roman period, such as, for example, the
Carrot Amphorae Camulodunum 189 and the Kingsholm
117 found in some important centres in the northern Adri-
atic and Po Valley,32 and they continued in middle Imperial
27
Within the different types of “one-handled jar”, more than just the
typical Ephesian forms are known. Moreover, productions from other
areas are also present in places such as Trieste (Auriemma 2007, 147)
and Aquilea (Villa 2013, no. 68).
28
González Cesteros – Yilmaz forthcoming.
29
Hayes 1991; Lund 2000b; Williams – Lund 2013; Bezeczky
2013b, 165-167.
30
Rauh – Slane 2000; Reynolds 2005.
31
Belotti 2008a.
32
Carre 2007, 591-592; Auriemma 2007, 150; Cipriano – Maz-
zochin 2011; Mazzochin 2013, 75; Dovreba – Riccato
18.03.19 14:39
 58 THE AMPHORAE
times with some occasional arrivals of forms like Chalk 6,
Agora K 108 and versions of baggy jars.33 Lastly, some Egyp-
tian AE 3 of the Roman period were imported to Trieste,34
but the presence of other Egyptian amphorae of the Roman
period is not documented in the region.
In addition to the above patterns, the northern Adriatic
region displays some individual characteristics concerning
oriental amphora types imported in the mid-Roman period.
One remarkable feature is the small incidence of one of the
most exported oriental types, namely the Kapitän 2.35 Even
if this form does appear in several places, their quantities are
always trivial, quite unlike the situation in other Mediterra-
nean regions, where it has been recovered in bigger amounts,
especially on the Aegean and the Black Sea.36
Another intriguing matter is that of the presence of the
so-called “Middle Roman Amphorae 8” in several northern
Adriatic centres. The bibliography about amphorae in north-
eastern Italy commonly alludes to the frequent presence of
this type during the late second and third centuries.37 Even
if it is undeniable that most of these pieces could be linked
with what Riley called MRA 8,38 his own definition of this
form is quite problematic. More attention to its postulated
production area in west Cyrenaica is recommended. Riley
mentioned the relevance of this form at the Tocra Kiln site,
based mainly on amphorae spikes.39 He also mentioned that
the pieces found in Berenice include one of the fabrics docu-
2015, 117. In the Augustan shipwreck of Cape Glavat of Mljet
Island, a probable Levantine amphora similar to the Augst 47 was
found, together with something similar to the Pamphilian amphorae
(Radić – Jurišić 1993; Jurišić 2000, Figs. 4, 8-9). We cannot say
anything more about these pieces at the moment.
33
Auriemma et al. 2012; Auriemma et al. 2015; Dovreba – Ric-
cato 2015, 117-118. The piece from the Aquileian well that appears
in: Degrassi – Maggi 2011, 265, Figs. 2,3, is not an Agora K 108,
but a baggy jar version of the third-fourth centuries A.D.
34
Auriemma 2007, Pl. 36, 100-102.
35
Verzár-Bass 1994, 401; Jurišić 2000, 21; Auriema – Quiri
2004, 52-53.
36
Within the Aegean, a very good example of the high quantities of
Kapitän 2 comes from several Ephesian contexts, dated to the end of
the second and the third centuries A.D. (Ladstätter 2005; Lad-
stätter 2008; Bezeczky 2013a, 149). Moreover, they have also
been found in huge numbers in almost every Mediterranean region
and along the Danubian frontier.
37
Ferrarini 1993; Belotti 2008b; Auriemma et al. 2012. In a visit
to the grain depot in Pompeii, we saw some amphorae that could also
be similar to the so-called MRA 8 and could confirm the beginning
of this production at least to the mid-first century A.D. However, this
observation is not certain and more research must be done on this
interesting topic.
38
Riley 1979, 193.
39
Riley 1979, 193.
BEZECZKY_Text.indd 58
mented at the Tocra workshop, as well as local fabric from
Berenice. Carre also pointed to the probable production of
this form in some kilns in Apollonia.40 Even if it presents
some resemblances with early forms of M 273, the MRA 8
seems to be a Cyrenaic production. The problem is that this
amphora type has not been found in any significant quantity
anywhere except at the northern Adriatic sites;41 or at least
they have not been recognized or published as MRA 8 any-
where else. The MRA 8 is well-represented in several places
in our region,42 including Trieste and Aquileia, always within
a chronological range from the mid-second to the mid-third
centuries A.D., a period for which we have textual and epi-
graphical evidence of the contacts between Aquileia and the
Cyrenaica.43
Before passing to the Late Antiquity phase, there is also
the important question about the contents of the east-
ern Mediterranean amphorae found in the whole Adriatic
region during the early and middle Roman times. Most of
the amphora types of these manufacturing contexts are
linked to wine consumption, and held such well-known
wines as the Rhodian, Koan, Knidian, Chian or the Ephe-
sian brands. However other kinds of goods produced in the
eastern regions were so transported too. Together with some
dried fruits or dates carried in the Levantine Carrot Ampho-
rae and Kingsholm 117,44 we also find some alum ampho-
rae from Milos. Moreover, the high number of Dressel 24
and related types, mainly in the second and third centuries’
versions, as found in Trieste and other northern places, are
significant too.45 These forms are connected to the olive oil
export of the Aegean, as the finds in Monte Testaccio and
the inscriptions mentioning oil have already confirmed.46
A titulus pictus on the neck of a Dressel 24 (Knossos 15)
40
Carre 2007, 596.
41
As an example, their presence in the Roman baths of Leptis Magna
in Tripolitania is reduced to the level of some sherds (Bonifay –
Capelli 2013, Tab. II).
42
Belotti 2008b, 455; Auriemma et al. 2012.
43
Belotti 2008b, 456; Auriemma et al. 2016, 393.
44
Auriemma 2007, 150; Cipriano – Mazzochin 2011; Mazzo-
chin 2013, 75.
45
In Aquileia, the Dressel 24 are well-documented too (Auriemma
2007, 142-145; Auriemma et al. 2012, 172; Auriemma et al. 2016),
as well as in Oderzo (Cipriano – Ferrarini 2001). We should also
mention again the presence of two early Dressel 24 types within the
cargo of the first century Cape Glavat shipwreck (Radić – Jurišić
1993, 125). For a complete view of the diffusion of this form in the
region: Belotti 2008b.
46
Their own presence in the Testaccio, as an olive oil amphorae deposit,
as well as some tituli picti found there and mentioning oil as the con-
tent (Carreras Monfort 1999, 97-98; Remesal Rodríguez –
García Sánchez 2007, 173-182), support this hypothesis.
18.03.19 14:39
 CHAPTER 6 – Eastern Mediterranean amphorae (González Cesteros)
from Aquileia recently published by Auriemma and other
colleagues,47 indicates liquamen ma… as the content.48 Even
if an occasional export of fish products in Dressel 24 cannot
be rejected, and some scholars have argued for this use based
on both the typological division between Knossos 18 and
Knossos 15 and some tituli picti mentioning supposed fish
sauce content,49 we believe a secondary use of this particular
piece is a more plausible option. The use of Latin and not
Greek in the titulus, and a comparison with the amphorae
from the Grado shipwreck,50 where wine amphorae were
reused and contained a fish product, points in this direction.
We agree with Belotti51 and believe we must assume that
olive oil was the content of most of the Dressel 24 exported
to the northern Adriatic, even if some exceptions existed.
In this volume and in other publications, the theories that
linked the abrupt decline of the Istrian amphora produc-
tion and export with a tragic cessation of the Istrian olive
oil production have been set out.52 In the pages dedicated
to the Hispanic imports, we associate the arrival of Dressel
20 during the Antonine period with the halting of the mas-
sive Istrian olive oil production. Nevertheless, we have not
documented the presence of Dressel 24 and related forms in
Istrian contexts outside the Trieste area, neither is it the case
at the Castrum and in the other excavations on the Brijuni
islands. But equally no fragments of Cyrenaic MRA 8, also a
possible olive oil container that arrived in the northern Adri-
atic region during the mid-second and mid-third centuries
have turned up. We must assume that within a few years after
the cessation of the Dressel 6B production the Istrian olive
oil was again able to cover local demand. Thus it is prob-
ably to be expected that only small amounts of foreign olive
oil amphorae were imported, and that, too, into the most
important settlements, where indeed some Dressel 20 have
47
Auriemma et al. 2016, 392-393.
48
Auriemma et al. 2016 published the titulus as liquamen mattum,
which seems to us a little bit strange, since mattum means humid or
impregnated and liquamen is already a liquid product. It could be a
kind of mixed liquamen, impregnated with oil, wine or any other liq-
uid. Another possibility is to read maturum and not mattum, being
related to the quality of the fish product, but this does not seem
exactly what is written according to the little epigraphical evidence
existing on this point. However, we have not seen the piece and must
believe in the author’s reading. We would like to thank E. García Var-
gas for comments on this subject.
49
Auriemma et al. 2016, 392-393; Dyczek 2008, 517-518, based on
tituli picti, proposed a sporadic use of Dressel 24 for other contents.
50
Auriemma 2000.
51
Belotti 2008.
52
Bezeczky 1998a, 10-11.
BEZECZKY_Text.indd 59
59
been documented,53 probably to cover the above-average
demand such larger places generated. On the other hand,
the Istrian olive oil production must have been unable to
supply the ancient markets associated with Dressel 6B any-
more, neither the nearest, such as Trieste, nor the important
cities of Po valley, even though, in the recent past, the small
containers that continue the Istrian production during the
second part of the second century and the beginning of the
third century A.D. have appeared in small quantities in sev-
eral places in northern Italy.54
3 LATE ROMAN AND EARLY BYZANTINE
PERIOD
The circulation of these above amphora forms ceased in
Late Antiquity, when new types appear in our region. As
we have already seen, the presence of eastern forms is quite
significant in early and middle Roman times and this situa-
tion continues during Late Antiquity, indeed it is now that
they reach their peak, providing the majority of the over-
all amphora ensemble in several contexts. These imports
arrived in huge numbers from the late fourth till the seventh
centuries in the whole Adriatic region. Even so, during the
fourth and early fifth centuries the high presence of Afri-
can imports, quite important in places such as Trieste and
its Slovenian hinterland,55 Aquileia56 or even in Ravenna,57
kept even quantities of oriental imports in second place, or
sharing the first. It is from the second part of the fifth cen-
tury till the seventh century or even later, due mainly to a
massive presence of LRA 1B and LRA 4, when the produc-
tions from the eastern Mediterranean really take over, even
penetrating into inland places in Friuli and Slovenia.58
In Late Antiquity, we discover almost every kind of east-
ern Mediterranean amphora in the Adriatic,59 but there are
53
See González Cesteros – Berni Millet in this volume.
54
Auriemma 2007, 129-130. We must keep in mind the limited
capacity of these containers. Archaeologists are not always in agree-
ment about the content of such forms; even if they continue the
typological line marked by the Dressel 6B, some tituli picti seem to
refer to fish products.
55
Auriemma – Quiri 2004; Auriemma 2007; Vidrih Perko –
Župančić 2005; Vidrih Perko 2006.
56
Auriemma 2007, 152; Villa 2013.
57
Augenti – Cirelli 2010.
58
Villa 1998; Vidrih Perko 2006; Auriemma – Quiri 2007.
59
If we compare the relatively good numbers of Cretan amphora types
that arrived during the Roman period, their near absence now is the
only remarkable change apropos the Oriental regions exporting their
products to the northern Adriatic. It must probably be explained as
part of the economic and commercial changes of Late Antiquity,
18.03.19 14:39
 60 THE AMPHORAE
also changes in the main oriental suppliers to the northern
Adriatic consumers. Cilicia-Cyprus and the Levant have by
now already displaced the Aegean basin as the bulk import-
ers. The colossal import numbers of LRA 1A and the even
bigger quantities of the later 1B version,60 followed by dif-
ferent versions of the Palestinian LRA 4, demonstrate the
special link that existed between the Adriatic and these two
regions. This is further underlined by the fact that these two
amphora types in all their different variants have been found
in almost every place in our region.61
The LRA 1 seems to have been the only Cilician-Cypriot
import in our area in Late Antiquity, but from the Levan-
tine coastal region, together with the Gaza containers, other
kinds of amphorae produced in today’s Israel and Lebanon
also arrived. In some coastal places, with contexts dated in
the fourth and fifth centuries, above all in Ravenna and Aqui-
leia, the form Agora M 334 is present in limited quantities,62
alongside other types linked with the Carrot amphorae tra-
dition. These containers must have been produced in Phoe-
nicia, the M 334 above all in the Akko region.63
Maybe some LRA 4 had arrived earlier, but their presence
seems to be important only from the fifth century onward.
One must observe the high quantities of LRA 4 found in
Aquileia64 and above all in Ravenna.65 The relevance of
Ravenna as the Imperial capital and the main civilian, reli-
gious and administrative centre from the early fifth to the
eighth centuries is well known: it makes Ravenna an advan-
tageous point for the study of the commercial trends that
took place in these centuries. Excavations carried on in the
last decades in Classe, the main harbour of Ravenna, only a
few kilometres from the city, brought to light huge quantities
of pottery from this period.66 In some late fifth and sixth cen-
tury contexts in Classe, large numbers of LRA 5/6 ampho-
when the distribution of Cretan amphorae is mostly limited to the
eastern regions.
60
In Aquileia Auriemma and Quiri proposed some years ago a much
bigger presence of the earlier LRA 1A than the later LRA 1B
(Auriema – Quiri 2007, 38), and this view has been reinforced by
the latest excavations in the city (Dobreva – Riccato 2015, 115-
117). The cause for this situation must be linked to the decline in
importance of this city during the late fifth and sixth centuries: newly
published material underlines the decline of imports from the mid-
late fifth century, even if eastern Mediterranean and African ampho-
rae do continue to arrive (Ceazzi – Del Brusco 2014).
61
Auriemma – Quiri 2007.
62
Stoppioni Piccoli 1983, 132-133; Cirelli – Cavanicci 2014;
Auriemma 2007, 147; Dovreba – Riccato 2015, 121.
63
Reynolds 2005.
64
Dovreba – Riccato 2015, 119.
65
Augenti – Cirelli 2010.
66
Augenti – Cirelli 2010; Cirelli 2014.
BEZECZKY_Text.indd 60
rae have been discovered,67 but in other contexts in Ravenna
and in other places in the northern Adriatic they seem to
be just sporadic imports, with a few being documented in
Aquileia,68 maybe in Trieste,69 certainly in its surrounding
region (as for example in Koper),70 and more thinly in Fri-
uli71 and in the military sites of Slovenia.72
Although the Levant and Cilicia-Cyprus were the main
oriental export regions in the first phases of Late Antiquity,
by the fourth and early fifth centuries we can also see the
presence of different Aegean forms in the northern Adriatic.
Even if already present on the Dalmatian coast, a complete
piece of the form Agora M 235/6 deserves attention: it was
found in the Wall depot of Salona, whose construction dates
from the Theodosian period of government,73 but unfor-
tunately no more pieces of this form have turned up in the
northern areas,74 with the exception of one exemplar found
in Brijuni. A possible Cretan origin has been proposed for
this form,75 something that seems possible to us. But our
personal experience in Ephesus, where the form is present
within the harbour necropolis finds76 in different fabrics,
leads us to hypothesise that it is still early to support this sole
Cretan origin, but rather that the possibility of more than
one production site must also be considered.77
From the late fourth to the end of the sixth centuries, the
most widespread Aegean containers were the small LRA 3
produced in the Cayster valley and nearby areas.78 The dif-
fusion of the different versions of these small amphorae
reached its peak in the fifth and early sixth centuries, being
retrieved in almost every place in the Mediterranean basin
and even outside this region. The northern Adriatic is not an
67
Augenti – Cirelli 2010, 609.
68
Vezár-Bass 1994, 411; Auriemma – Quiri 2007, 47; Ceazzi –
Del Brusco 2014; Dovreba – Riccato 2015, 119-121;
Dovreba – Riccato 2015, 119-121. As mentioned before, an early
version of baggy-jar amphorae dated in the third-fourth century was
found in a well in the Forum of Aquilea (Degrassi – Maggi 2011).
69
Auriemma 2007, 150.
70
Cunja 1996.
71
Auriemma – Quiri 2007, 47.
72
Vidrih Perko 2006, 216; Auriemma – Quiri 2007, 47.
73
Cambi 1989, 331-335; Mardešić 1994, 294-295.
74
These amphorae are also present in some southern Adriatic contexts,
as in Puglia (Volpe et al. 2010, 645-646, Figs. 5, 3-4), and it is pos-
sible they have been mixed up with other types, mainly LRA 2, in
other places.
75
Bonifay – Capelli 2013, 74-75.
76
González Cesteros – Yilmaz forthcoming.
77
M. Bonifay (personal communication) mentions a possible produc-
tion site in the Peloponnese. This possibility must also be kept in
mind, and we hope future discoveries will resolve the geographical
origin of this form.
78
Bezeczky 2013a.
18.03.19 14:39
 CHAPTER 6 – Eastern Mediterranean amphorae (González Cesteros)
exception, and they are commonly found there, which agrees
with the already mentioned relatively common presence of
their “one-handled” predecessors (vide supra). Together with
the LRA 3 in the sixth century, another type produced in the
Ephesian hinterland arrived in a few places in the northern
Adriatic. The so-called Ephesus 56 is to be found in Trieste,
where it has been confused with the LRA 3,79 and we have
also documented several exemplars within our material in
Brijuni. Their absence in contexts in Ravenna is strange, but
we have to admit that they could have been muddled up with
the more common LRA 3.
Similar fabrics are also seen in some Agora M 273 ampho-
rae, and mostly with the latest versions of this type named
the Samos Cistern amphorae. These two connected forms
are represented in several places in our region, but always
in small quantities, especially so for the earlier M 273.80 It
seems that the Samos Cistern form achieved more success
within the Adriatic,81 where it can be found even in some
inland sites in Friuli,82 Trieste and its Slovenian hinterland83
and in Classe.84
Another similar or related type to the M 273 is the Adam-
sheck 22/Torone VII85, found only in a sporadic way in our
region.86 It seems to be produced above all during the fifth
century; the exact production place is still an open question.
It has been proposed that it might be the island of Icaria,
west of Samos, on the basis of some tituli picti.87 Even if the
evidence is not so clear, we have to keep in mind this hypoth-
esis. Meanwhile a match for the characteristic calcareous
fabric of these amphorae (with cream or whitish colour, low
quantities of mica and of small red inclusions, together with
occasional small rounded quartz grains) must be sought
both in future discoveries on this island, and elsewhere in the
eastern Aegean. Other production places cannot be ruled
out as some variety in the fabrics exists, and also the massive
79
See LR3 and Ephesus 56 in Bezeczky 2013a, 164-170; Degrassi
et al. 2007, Fig. 1.8. Even if the spike of this exemplar is not of the
usual sort, we must admit that this complete piece is an Ephesus 56
and not a LRA 3.
80
Vezár-Bass 1994, Pl. 71; Auriemma 2007, 145-146; Auriemma –
Quiri 2007, 44.
81
Auriemma – Quiri 2007, 44-45.
82
Arthur 1990; Villa 1998.
83
Degrassi et al. 2010; Vidrih Perko – Župančić 2005.
84
Stoppioni Piccoli 1983; Augenti – Cirelli 2010, 609.
85
Adamsheck 1979; Papadopoulos 1991.
86
Vidrih Perko – Župančić 2005; Cirelli 2014, Fig. 5.
87
Arthur 1998.
BEZECZKY_Text.indd 61
61
production that the form attained in the fifth century might
be more than a small island could achieve.88
The presence of LRA 2 in the region is one of marked
contrasts. The arrival of this form maintains the level of the
relatively well-represented Dressel 24 variants of the late sec-
ond and third centuries. However their numbers in some
main centres of the region, such as Ravenna89 or Aquileia,
are not particularly elevated,90 and only in Trieste does their
number reach a higher volume.91 Nevertheless they seem to
be more concentrated on the Slovenian route to Ljubljana,92
in the military places of Friuli93 or in Istria, where they have
been found in Koper94 in similar quantities as in Trieste.
Certainly they are present in good numbers in our Castrum,
where they form one of the most important amphora types
within the Late Antique material. Even if in some cases it
is possible that some seventh and eighth century globular
amphorae have unfortunately been classified together with
the LRA 2 in Slovenia, Istria or Friuli, the quantities of LRA
2 that arrived from the late fourth to the late sixth centuries
A.D. are really momentous. It has been suggested, and seem-
ingly favourably within the scientific community, that the
LRA 2 were containers directly linked with the supply of
the military, on the basis of their extensive presence on the
lower Danube border.95 The same can certainly be claimed
for the cases of the early Byzantine military sites in the Adri-
atic region too: the big quantity of LRA 2 found in the Bri-
juni Castrum also supports this idea and, in our personal
opinion, indicates that the Castrum was used as a military
base during the early Byzantine period, even if the number
of soldiers based there was not too high96. We are not sure if
the LRA 2 were used as olive oil amphorae, but the fact that
they developed from the latest versions of Dressel 24 and
their rounded/globular size could indicate this function.
88
The huge amount of this type found in the excavations of the Har-
bour Necropolis in Ephesus (González Cesteros – Yilmaz
forthcoming), and the intended future analysis on numbers of them,
will be very useful in resolving this question.
89
Cirelli 2014.
90
Auriemma – Quiri 2007. This interpretation could change in the
future for Aquileia, as recent excavations on the via Bolonia brought
up an important number of LRA 2, in comparable numbers with
other eastern Mediterranean amphorae (Ceazzi – Del Brusco
2014).
91
Auriemma 2007; Ventura et al. 2014.
92
Vidrih Perko 2006.
93
Villa 1998.
94
Cunja 1996.
95
Karagiourgou 2001.
96
It is quite possible that it developed as a military fortress and should
have been part of the Claustra Alpium Iuliarum: Mlakar 1975-76;
Begović – Schrunk 2009; Begović – Schrunk 2011a.
18.03.19 14:39
 62 THE AMPHORAE
Their presence in our region, as well as of some large cylin-
drical African containers, can also be helpful in determining
the nature of the challenge that the local olive oil production
faced in external imports during Late Antiquity.
From the fourth century, the amphorae from the Black
Sea increased their production levels, especially as the region
of Sinope started to convey more amphorae to its custom-
ers, mainly in the Black Sea and eastern Mediterranean. The
arrival of these products in the northern Adriatic region
seems to be virtually at the symbolic level only. We have
already mentioned the infrequent presence of Black Sea
products during the early and middle Roman periods, and
even if we can find some more exemplars from the fourth to
the sixth centuries,97 they never attain any real consequence
in the total number of pieces at any archaeological context.98
Their absence, above all in Ravenna, is quite significant. It
typifies the almost complete absence of these amphorae in
the western and central Mediterranean, the Adriatic being
no exception.
Finally, in some Adriatic sites the presence of a small num-
ber of Egyptian amphorae is documented. They form quite a
rare group in our region, appearing almost as samplers, even
in big commercial places such as Ravenna, where some LRA
7 have been found.99 They have also been retrieved in other
places such as Ljubljana,100Aquileia,101 Trieste,102 in Friuli
and in some points in the Venetian lagoon,103 but always in
really small quantities. Other Egyptian forms of Late Antiq-
uity have not been documented, except for some fragments
of AET 3/Kellia 172 found in Aquileia.104
The effective absence of amphorae from Crete within the
Late Antique imports also needs pointing out. Even if it is
possible that some Cretan types have been misidentified, we
agree that the Cretan exports seem to have a most limited
geographical diffusion from the fourth century A.D. The
only documented exemplars are a rim fragment of a possible
97
Belotti 2008a.
98
In the literature we have been using, we have only documented a pos-
sible C-Sinop III in the upper layers of the excavations on the north-
ern side of the river port in Aquileia (Maggi – Urban 2001, Fig. 6).
However, we cannot affirm it is really a Black Sea amphora and the
authors did not mention anything about this piece in particular.
99
Stoppioni Piccoli 1983; Augenti – Cirelli 2010.
100
Vidrih Perko 2006.
101
Vezár-Bass 1994; Auriemma – Quiri 2007, 48; Ceazzi – Del
Brusco 2014.
102
Auriemma 2007, 151; Degrassi et al. 2010, 580.
103
Auriemma – Quiri 2007, 48.
104
Vezár-Bass 1994, Pl. 64 AO 24; Auriemma – Quiri 2007, 48.
BEZECZKY_Text.indd 62
TRC 2105 found in Trieste,106 together with our piece in Bri-
juni.107
Before leaving the amphora forms found in the northern
Adriatic, attention must be paid to two types whose ascrip-
tion to a production place in the eastern Mediterranean is
problematic. We are talking about the San Lorenzo 7, found
in several northern Adriatic and Italian settlements, but
always in small quantities,108 and the so-called “Vigna Bar-
berini 13”,109 an amphora present in late sixth century con-
texts in Ravenna.110
The San Lorenzo 7 turns up in the central and eastern
Mediterranean terrain; it is presumed to have been produced
on the Aegean from the second half of the second to the late
fourth or early fifth centuries.111 We are in broad agreement
with this postulation, that seems reinforced by some analy-
ses made in Milan,112 but due to the range of fabrics and the
extensive diffusion of this type in Italian contexts, we prefer
to leave the question open for future studies.
The second and more problematic type is the so-called
“Vigna Barberini 13” found only in Ravenna within the
Adriatic region. Cirelli has proposed an oriental origin for
this form, based on the abundant mica in the fabric and a
certain distribution weighted towards the eastern Mediter-
ranean. It is true that the type has resemblances with some
Late Antique Cretan amphorae, mainly with form TRC 1
of Portale and Romeo,113 but we think it has been confused
with the Cretan amphorae. For us it is a different form, much
closer to the south Italian and Sicilian Late Antique flat-bot-
tomed amphorae, such as, for example, the latest forms of
Keay 52 or the Crypta Balbi types. The fabrics of the Crypta
Balbi 2 amphorae are in fact quite similar to that com-
monly described for the “Vigna Barberini 13”, namely hard
and of red or brown-red colour and with abundant mica.
The petrological analysis carried out on some pieces found
in Rome supports the macroscopic visual identification.114
However, the latest analysis of “Vigna Barberini 13” pieces
found around the Palatine hill shows how difficult it is to
determine a concrete production region, as both the Aegean
105
Portale – Romeo 2000.
106
Degrassi et al. 2010, 580.
107
Vide infra.
108
Auriemma – Quiri 2007, 52; Dobreva 2012, 104-105.
109
Rizzo et al. 2004.
110
Augenti – Cirelli 2010, 609; Cirelli 2014, 543.
111
Arthur 1998, 171-172; Mongardi 2016.
112
Dobreva 2012, 104 on Lusuardi Sienna 1990, 384.
113
Portale – Romeo 2000, 422.
114
Pacetti 1998, 204; Casalini – Crespi 2010.
18.03.19 14:39
 CHAPTER 6 – Eastern Mediterranean amphorae (González Cesteros)
and southern Italy remain possibilities.115 Lastly, the actual
geographic distribution of our type is almost constrained to
Italy, being found above all in Rome and Ravenna, and in
Kellia within the eastern Mediterranean.116 In the Aegean
area we have recently documented a piece coming from the
ancient harbour excavations in Ephesos.117
Though but a footnote to the diffusion of Late Antique
eastern Mediterranean amphorae in the northern Adriatic
region, the above must be seen as part of a common trend
developing from the late fourth century A.D. that affected
the whole Mediterranean basin.118 Likewise, we believe that
the large quantities of amphorae from these regions found
in the Adriatic signify the continuous relations between our
region and the eastern Mediterranean. The continuity of
the same routes that connected the Adriatic and the Aegean
from the Bronze Age on must surely be admitted, as, too,
the increased volume of traffic from the fourth century A.D.
It is beyond doubt that the rise of Ravenna as the most
important urban centre in the whole region from the late
fourth century and above all from the fifth to seventh cen-
turies brought into being a new commercial infrastructure,
which finds its new main redeployment centre in the port
of Classe. Excavations carried on during the last decade
have highlighted the spectacular number of warehouses and
archaeological materials found in Classe, which can only be
understood by its having developed the role of an impor-
tant redistribution centre that probably supplied other main
points in Cisalpina, Venetia and Istria. It is also important
at this point to underline the special character of Ravenna,
now being the centre of the Eparch and court, as well as one
of the most important religious settlements with a power-
ful bishop. Even if well-represented along the Adriatic any-
way, the numbers and percentage of LRA 4 and of LRA 3 in
Ravenna contrast with other places in the northern Adriatic,
among them Brijuni, where the number of LRA 1 is over ten
times that of LRA 3 and LRA 4, with the LRA 2 taking sec-
ond place in the oriental amphorae represented. Places close
to Ravenna, such as Rimini, seem to present a similar but less
dramatic version of this pattern: here the LRA 1 is still the
most common type,119 or at least is present in a higher per-
115
Casalini – Crespi 2010, 104.
116
Egloff 1977, type 171.
117
For us the form found in Thessaloniki by Papanikola-Bakirtzi
2010 and identified by Cirelli 2014, 543 as the same form they have
in Ravenna is, in fact, something different with a rounded base and
different rim: it must be linked with the Cretan production, as with
forms TRC 1 or TRC 2 of Portale – Romeo 2000, 422.
118
Panella 1989; Reynolds 1993; Remollà 2000; Pieri 2005.
119
Negrelli 2008.
BEZECZKY_Text.indd 63
63
centage than in Ravenna itself. Another important centre in
the region in the fifth and sixth centuries looks to have been
Trieste, a main port from where products must have been
redistributed to its Slovenian hinterland and the western
Istrian coastal settlements, for example Brijuni.120
This special link between our region and the eastern Med-
iterranean territories continues beyond the chronological
frame being discussed here: we can still observe the arrival of
oriental products, again amphorae, to main centres such as
the Venetian lagoon or Comacchio from the late seventh to
the ninth or even tenth centuries.121 They are also found in
Croatian waters even later still.122
THE EASTERN MEDITERRANEAN AMPHORAE
TYPES FROM THE BRIJUNI CASTRUM VILLA
Considering the broad diffusion of the eastern Mediterra-
nean amphorae in the Adriatic from the Hellenistic period
till Late Antiquity, it is not surprising that a good quantity of
oriental amphorae have turned up within the ceramic mate-
rial of the Brijuni Castrum.123
It is obvious that the Castrum follows the trends discussed
for the complete northern Adriatic region, reminiscent of
places such as Aquileia, Trieste or Ravenna in the late Roman
and early Byzantine periods, but it also has some marked
peculiarities of its own. Here it should be observed that the
main eastern Mediterranean types are the most common
of the amphorae found in Castrum, such as the Rhodian,
Koan and Knidian during the Augustan and Julio-Claudian
period, or the LRA 1 in Late Antiquity. It is worthwhile,
though, to separate the oriental forms assignable to the early
and middle Roman period from the Late Antique imports,
because, even if during the Roman period their presence is
greater than other kinds of import from outside the Adriatic
region, it is in Late Antiquity that they really outstrip their
rivals on the island, even the African imports. This feature
has already been mentioned by Vidrih Perko and Pavletić.124
The phenomenon, without any doubt, occurs because the
biggest quantities of amphorae in the Castrum belong to the
late Roman and Byzantine occupational phases, a time when
120
We have already pointed out the value of a precise study of the ampho-
rae from the Roman and Byzantine periods in Pula: comparing it with
other main places along the northern Adriatic, above all with Trieste
and Ravenna, would probably be most instructive.
121
Gelichi – Negrelli 2008.
122
Brusić 1976; Brusić 2010; Zmaić 2013.
123
Bezeczky et al. 2015.
124
Vidrih Perko – Pavletić 2000.
18.03.19 14:39
 64 THE AMPHORAE
a military unit was probably accommodated in the Castrum
and a religious community in the nearby church.125
The late Roman and Byzantine period is a time when
the Adriatic territories do not seem to have exported their
surpluses in amphorae, as opposed to Roman times when
Istria was an amphora production region. Even if amphorae
were not produced on the island (we have not yet found any
remains of kilns and debris linked to pottery production),
it is certain that olive oil and other commodities were pro-
duced there.126 On the other hand, amphora production is
attested in the workshop of Fazana during early Imperial
times, which lies just on the other side of the Brijuni channel
and is one of the natural ports for the archipelago.127
Within the oriental material of the Castrum we find
material from all kinds of provenances and products, with
the greatest import of the Imperial period belonging to wine
amphorae from the Aegean region, while in Late Antiquity
it is now the wine from Cilicia-Cyprus, from the Aegean and
Levant that represents the bulk of the imports. The absence
of oriental amphorae linked with olive oil during the Roman
period is noteworthy, though the LRA 2, from the eastern
Mediterranean with its possible olive oil contents, is repre-
sented in the Castrum during the Late Antiquity. This pat-
tern contrasts with that of other places in the region, mainly
Trieste and Aquileia, where amphorae such as the Dressel 24
and related forms, or the MRA 8, also a possible olive oil
container, are commonly encountered.
Due to the stratigraphical problems inherited from the
earlier unprofessional excavations that took place in the Bri-
juni Castrum,128 no clear distinction between the material of
the early and later periods is feasible. One must rely directly
on the typological distinctions for a better chronological
understanding. One would, however, expect an increased
presence of amphora imports during the latest occupational
phase, as the so-called Castrum acquired a different role
at this time, transforming it from a productive pars rustica
of a Roman villa to a military fortress.129 The local histori-
cal development of Istria in Roman times made the whole
region into an olive oil production and exporting area,
with good established markets in northern Italy and central
Europe;130 it exported many more amphorae and goods than
it brought in in balancing or luxury products. This situa-
125
See Schobert in this volume.
126
Bezeczky 1998a; and Bezeczky (Chapter 4) in this volume.
127
Gnirs 1910; Gnirs 1911b; Bulić – Koncani Uhač 2011.
128
Vidrih Perko – Pavletić 2000.
129
See Chapter 3.1.
130
Bezeczky 1998a; and Bezeczky (Chapter 4) in this volume.
BEZECZKY_Text.indd 64
tion, though, changed from the mid-second century A.D.:
a higher amount of imported produce is to be expected
from this time on, something that is confirmed for the Late
Roman times.
EARLY AND MIDDLE ROMAN PERIOD
The material from the Castrum includes amphorae coming
from the eastern Mediterranean, running from the late first
century B.C. to the early seventh century A.D. As already
mentioned, some of this attained a very high profile in the
Late Roman and Byzantine phases, even though of minimal
status beforehand; moreover, typical forms for the second
and third centuries A.D. are still only found in quite small
quantities. Eastern products, however, arrived here continu-
ously during Roman Imperial times, from many different
regions, although the most represented are the well-known
Aegean Imperial productions.
From the Augustan period, the arrival of eastern Mediter-
ranean amphorae at the Castrum is observed. Among them,
the most important types produced in the Aegean basin at
this time are present, as they are in other places in the north-
ern Adriatic. However, a key difference exists with elsewhere
in northern Italy, Slovenia and Croatia: namely the “ele-
vated” number of Koan and Knidian amphorae compared to
the Rhodian,131 a pattern completely different from that seen
in other places, above all in the Cisalpina. Maybe this differ-
ence is only due to chance, but equally the dynamic of the
Castrum and the Brijuni islands may not have been exactly
the same as in other places.
The presence of a small Augustan “one-handled jar” of
form Agora F 65/66 within our material has been docu-
mented. Writers of early Imperial times mention the quality
of the wine produced in Ephesus and nearby places,132 where
this vase form was normally produced, but in our exemplar
the fabric seems not to be Ephesian but rather from some
other place in Asia Minor or the nearby islands, where these
containers and their later versions with two handles were
also produced. The absence of other oriental amphorae
linked with high quality wines for this period, above all of
Chian amphorae, argues against any luxury imports in the
131
We have to remember that the typology of some of the Rhodian
amphorae found in Brijuni places them in the second half of the first
century A.D.
132
Ladstätter 2008, 181-182; Bezeczky 2013a, 25.
18.03.19 14:39
 CHAPTER 6 – Eastern Mediterranean amphorae (González Cesteros)
Castrum,133 though maybe they are located in other places
on the island, like the luxury villa of Val Catena.
The later versions of Koan amphorae, frequently known
as Dressel 5, are present in Brijuni, pointing to the continu-
ous arrival of these wines during the end of the first and the
second centuries A.D. However, not all the fabrics found in
these pieces can be assigned to the island of Kos – a broad
Aegean and even Pamphilian and Cilician production is to
be expected. Equally, some of the Rhodian fragments seem
to be late versions, probably produced during the second
part of the first century A.D., as, for example, the catalogue
entries (58 and 59).
Other products from the Aegean and nearby areas are
present too. A few fragments of AC 4, a typical Cretan
form based on the Rhodian types of the first and early sec-
ond centuries,134 have been found in Brijuni. It is not always
easy to distinguish this kind of amphora from the Rhodian
Imperial amphorae, especially in cases of big modules. The
handle fragment (69) corresponds to a small module with
a really prominent horn. It can be classified without any
doubt among the small-size AC4 produced in the late first
century A.D.
The arrival of other oriental containers, for example some
“one-handled” vessels from Ephesus (forms M 125 and M
126), and also the handle fragment (70) of Agora G 199 that
presents a rough, soft, probably Cilician, fabric, can also be
dated to the late first and second centuries A.D. These forms
were produced over a longer period: they are normally to be
found in second and early third centuries contexts too.
Now, Cretan products continued to arrive at Brijuni, but
just sporadically: a couple of fragments of form MRC 3 of
Portale and Romeo (103). But it is again the “one-handled”
Asian container that constitutes the most important import
of the late second and third centuries. Some different ver-
sions of these small containers, above all the form Agora M
240 with the typical Ephesian fabric, are known. The pres-
ence of an Agora M 279 similis with another kind of fabric
without mica requires comment: this must be from a manu-
facturing locale that excludes big areas of Asia Minor, such as
Ephesus and the Meander valley.
A base (78) of type Kapitän 1 and some rims and han-
dle fragments of Kapitän 2 were recovered. The fabric of
133
Based on the epigraphic evidence of Delos in the second century B.C.,
Tchernia 1986, 105, mentions that the Knidian wines were four
times as expensive as the Koan. We cannot be sure that the wines of
Knidos were high end wines, but it must be admitted that they were
not usual wines.
134
Marangou 1995, 84.
BEZECZKY_Text.indd 65
65
the Kapitän 1 seems to be of northern Asia Minor origin,
or from the nearby islands like Chios or Lesbos. It is red in
colour, medium-hard and laminated with small quantities of
mica and other small stone inclusions. On the other hand,
the Kapitän 2 fragments are of two different fabrics, one of
them is the most typical, that is brick red in colour, hard and
coarse with big grey and white inclusions, while the second is
smoother and finer, with temper smaller-sized and lower in
quantity (79). As already considered, almost any place in the
northern Adriatic region has yielded big quantities of these
two forms. While the situation is more or less normal for the
Kapitän 1, in case of the Kapitän 2 it must be regarded as
an exception within the broad dispersal of this type seen in
most of the Mediterranean regions.
LATE ROMAN AND EARLY BYZANTINE PERIOD
It is in this period, between the fourth and early seventh cen-
turies A.D., when the products coming from the eastern part
of the Mediterranean reached their zenith, that they and the
African imports constituted the bulk of the amphorae found
in the Castrum. These significantly increased quantities of
oriental amphorae must find their explanation in the pres-
ence of soldiers and the church personnel in Brijuni.
Such quantities of African and oriental amphorae are
usually believed to reflect the Byzantine occupation of the
central and western Mediterranean carried out by Justinian
during the central decades of the sixth century. In our opin-
ion, this is an extremely simplistic explanation of a broader
and more complex economic process that requires a more
nuanced and multiple approach: not only will the Byzantine
presence in Italy or Africa have played a vital role, but also
the growth in output of northern African produce and that
in most of the eastern Mediterranean territories,135 not to
mention the close commercial links that were always practi-
cable between the Aegean and the Adriatic.
One important change that took place apropos the Late
Antique eastern types is the replacement at that time of the
Aegean basin by the Cilician and Cypriot areas as the most
important source regions supplying amphora products to
the Castrum. It is the same process already remarked upon
and that involved the whole northern Adriatic area, where
the Cilician and Levantine products always hold premier
place among the imports of this phase. The sheer volume of
LRA 1 pieces makes the relevance of their imports for the
135
Bonifay 2004; Pieri 2005.
18.03.19 14:39
 66 THE AMPHORAE
Adriatic region136 starkly obvious here, just as it does for the
whole Mediterranean area and beyond. The LRA 1 in fact
achieve a much higher percentage in Brijuni than in any
other place studied in our region, accounting for more than
20 % of the total amphorae. This percentage is similar to that
determined by Vidrih Perko and Pavletić in their prelimi-
nary study.137 The most probable explanation exists, once
again, in the unique nature of this settlement during Late
Antiquity.138
Most of the LRA 1 belong to the typical forms of the
late fifth and sixth centuries and should be placed within
the form LRA 1B of Pieri.139 This in itself is nothing new,
because, firstly most of the amphorae imports are placed in
the late fourth to early seventh centuries and, secondly, form
1B was the variant of the LRA 1 that got exported most. The
fabric of these pieces is typical of material from the Cilician
and Cypriot areas, presenting different colours, but being
hard or medium-hard, sandy or coarse with a number of
white and yellowish limestone pieces, as well as other grey
and sometimes red inclusions. There is one amphora rim
and handle (85) which could come from the Soli workshop
(Turkey), with its calcareous fabric.140
The LRA 2 hold a quite elevated position within the
Castrum material (87-88), apparently on a scale greater
even than in Trieste or Koper. The argued link between this
particular form and the army supply as documented on the
lower Danube border141 could also apply for the Castrum
and the northwestern Adriatic territory. The only problem
we find with this hypothesis is that of the presumed con-
tent of these containers. If it was olive oil, which probably
it was at least in part, then it raises the question of what was
happening to the local Istrian production at this time. Was
Istria not able even to supply the soldiers and military staff
settled on its borders anymore? There is no easy answer to
this question. The remains of three Late Antique oil-presses
inside the Castrum,142 be they used for oil, wine or both,
do not help us in coming to a better understanding of the
136
Cirelli 2007, 47.
137
Vidrih Perko – Pavletić 2000.
138
It has been suggested that there is a connection between some
amphora types such as LRA 2 (Karagiourgou 2001) or LRA 13
(Diamanti 2010a; Diamanti 2010b) and the needs of the early
Byzantine State, as is expressed in the annona, whether military or
civilian. It could account for the elevated number of LRA 1, but the
question is difficult to answer since this is the case wherever one looks
in the Adriatic.
139
Pieri 2005, 75-77.
140
Autret et al. 2010, 203-207.
141
Karagiorgou 2001.
142
Matijašić 1993a, 249-251.
BEZECZKY_Text.indd 66
elevated quantities of amphora imports from this period.
Most of the LRA 2 pieces belong to the latest variant docu-
mented in the late fifth and sixth centuries: they have taller
necks and handles, not really rounded bodies and have wavy
incised lines in the upper part of the body walls. Neverthe-
less a good number of pieces from the earlier versions of the
late fourth and early fifth centuries were recovered too. We
also have to point out that within our material not a single
globular amphora from the late seventh century and later has
turned up, although we do have the presence of an LRA 13
with an Imperial stamp (100) and some late African forms
documented. The fabric of most of the LRA 2 fragments
is hard, a coarse reddish-brown on the external surface and
more yellowish on the inner face. It presents some white
limestone inclusions, sometimes large, and also some voids
of different sizes. The source of this fabric must be placed in
the Argolid and Boeotia, where some kiln sites have already
been found.143 However, some pieces with a finer fabric
occur, with small or medium white, red and occasional grey
inclusions: these probably come from Chios or the nearby
territory in Asia Minor.144
A special piece is no. 104. It is an almost complete rim
fragment, with part of the neck and a handle, of the form
Agora M 235/6, a type that recalls the LRA 2 in shape. We
have already mentioned the small presence of this fourth and
early fifth century form in the Adriatic: the piece at Brijuni
seems almost exceptional. The fabric of this exemplar is light
red, hard and fine, sandy, with some red, grey and white
quartz inclusions. We are not able to affirm the provenance
and the continent of this amphora, but we propose an origin
in Crete or on the Greek mainland.145
Within the same family of containers belongs a neck
fragment with an incomplete stamp of a Byzantine Emperor
(100), for which the best-known parallels are similar stamps
found in Alexandria,146 Istanbul Saraçhane and Yenikapı,147
Athens,148 Geronissos island (Cyprus),149 Kellia,150 Tocra151
and the island of Kos,152 where this amphora form was pro-
duced.153 With the publication of the discoveries from the
143
Rudolph 1979; Gerousi 2014.
144
Opait – Tsaravopoulos 2011; Arthur 1990, 286.
145
Vide supra.
146
Sztetyłło – Borkowsky 1986.
147
Hayes 1992; Kara 2015.
148
Grace 1949; Opait – Diamanti 2014.
149
Connely – Wilson 2002, 274-275.
150
Bonnet – Cattin 1999, 538-539.
151
Bentaher 1994, 231.
152
Diamanti 2010a; Diamanti 2010b; Diamanti 2012.
153
Also produced in Paros: Diamanti 2016.
18.03.19 14:39
 CHAPTER 6 – Eastern Mediterranean amphorae (González Cesteros)
Halasarna workshop on the eastern coast of Kos, the number
of stamped pieces documented has significantly increased.
Our stamp presents the bust of an emperor holding a sceptre
in the left hand and probably the item of clothing called a
mappa in the right hand. Around the emperor’s bust a par-
tially preserved inscription is to be seen. We can distinguish
some Greek capital letters in the lower right part of the stamp
and maybe in the upper left too, but in this last position they
have been effectively erased. These letters can be read as: (--)
ΟΛ(------) and must correspond with ΠΤΟΛΑΙΜΕΟΥ.
The best parallel examples are two stamps from Athens,
recently published by Opait and Diamanti.154 The Athenian
exemplars preserve more of the stamps and were stamped on
the neck too. The authors mention the similarities with oth-
ers stamps found in Kos and date them to the last decades of
the sixth century,155 before the reign of Phocas, who appears
on stamps too, but is always represented with a large beard,
which is apparently not the case in our piece. This piece, clas-
sified as an LRA 13, a term not without controversy,156 seems
to correspond with a product of the well-known kiln site of
Halasarna (modern Kardamaina). Both the epigraphic evi-
dence and the visual characteristics of the fabric (red-brown,
hard, fine and micaceous, with dark grey and quartz inclu-
sions) correspond to the description given by Diamanti
about the production of this and other kinds of amphorae in
the Halasarna workshop157 much more closely than to items
produced in other Late Antique workshops in Kos described
by Didioumi.158
In the northern Adriatic region, in excavations in Byzan-
tine Comacchio, a seventh or early eighth century globular
amphora produced in some Aegean workshop was recov-
ered: it bears a circular stamp with two letters on the neck.159
Although the stamp on this piece probably serves the same
purpose as that mentioned above, we think the different
154
Opait – Diamanti 2014, 55, Figs. 4 and 9.
155
Opait – Diamanti 2014, 55.
156
The problem about this denomination comes from the Berenice publi-
cation. Riley argued that within this type there was a broader variation
from the late sixth and seventh centuries, which drew its descent from
both the sixth century LRA 1 and LRA 2. In some publications we
find the term LRA 13 used for the late globular amphorae (Demes-
tica 2005), as well as for smaller and thinner containers derived from
the LRA 1 (Diamanti 2010). It is true that this late form seems to
have taken some characteristics of both LRA 1 and LRA 2 types, but
it would be more appropriate to make a clear division between the two
types (we would like to thank Dr. Diamanti for the opportune conver-
sation about these types during the Rei Cretariae Congress 2014).
157
Diamanti 2010a; Diamanti 2010b, 1.
158
Didioumi 2014.
159
Gelichi – Negrelli 2008, Fig. 11.1.
BEZECZKY_Text.indd 67
67
type of amphora and the different form of the stamp suggest
a different production place. According to Diamanti, stamps
of this kind are only to be found on amphorae produced in
the late sixth and early seventh centuries. They represent
some kind of state control linked with both production and
commercial activities, probably under Imperial administra-
tion through the control of officers linked to the Eparch:
perhaps the Kommerkiarioi, a group of public officers with
different economic duties – for example being in charge of
taxes, or controlling the imports and exports with other ter-
ritories.160 From the late sixth to the early ninth centuries,
the Kommerkiarioi had stamps with the bust or silhouette of
the emperor (as is discerned in our fragmentary example),161
which they used to certify the good quality and guarantee
the commercial trafficking of the stamped products.162
This fragment is a good example of the continuity of
Aegean imports into Brijuni. Within the Aegean, a simi-
lar range of production centres and amphora types may
be observed in early and middle Roman times, but now
the most important places are no longer the islands of
Kos, Rhodes and nearby coastal territories in Asia Minor.
Despite the stamped amphora fragment already mentioned,
a Dodecanesian provenance for any more Late Antique
amphora fragments, as based on the visual characteristics of
the fabrics, cannot be vouchsafed: even if some Samos Cis-
tern pieces seem to have similar fabrics to some early Roman
Koan amphorae.163 However, other territories in the eastern
Aegean, such as the central coastal band of Asia Minor, espe-
cially the territory around Ephesus and the nearby islands,
and also Chios and the continental territories close to it, all
seem to be well-represented within the Castrum assemblage.
The small wine containers produced in Ephesus and other
places in western Asia Minor and on some nearby islands
continued to arrive in the form of LRA 3 to Brijuni during
Late Antiquity.164 Within our material are a good number of
these small containers in their different versions and differ-
ent sizes, but always with the typical dark brown or a reddish
soapy and highly micaceous Ephesian fabric (89-90). They
are the third oriental type, but far fewer in number than the
LRA 1 and LRA 2. The relevance of the Ephesus hinterland
160
Diamanti 2010a; Opait – Diamanti 2014, 55.
161
Probably some late sixth, or better early seventh-century emperor:
Maurice (582-602), Phocas I (602-610), Heraclius (610-641).
162
Oikonomidès 1989; Diamanti 2010a.
163
Vide infra.
164
Imitations of LRA 3 and Ephesus 56 have recently been found on the
island of Kos, in the kiln site of Mastichari (Didioumi 2014; Bezec-
zky 2013a, 167).
18.03.19 14:39
 68 THE AMPHORAE
in the supply of the Castrum goes beyond the presence of the
LRA 3 and is enhanced by a goodly quantity of that other
important Late Antique Ephesian form, the so-called Ephe-
sus 56 amphora (91-92).165 We spoke in the first part of this
article of the remarkable absence of this form in the north-
ern Adriatic region, being only documented by one complete
exemplar from Trieste,166 where it has been wrongly classified
within the LRA 3. Even if amphorae of this form have been
misjudged and put in with the more diffused LRA 3, we still
believe the relatively elevated number of Ephesus 56 found
in Brijuni is a true exception – again to be connected with
the possible military unit settled there and the ecclesiastical
community of the church.
From the eastern Aegean region another amphora form
came to Castrum during late Roman and Byzantine times.
The Agora M 273 produced during the fourth and fifth cen-
turies A.D. is well-represented here, as well as their latest
version, the so-called Samos Cistern amphora, typical of
the sixth and early seventh centuries.167 Within our material
a high number of M 273 and related forms were counted,
with different fabrics that suggest various production places
within the Aegean area. Some of them show mica and are
a dark brown-reddish colour, similar to the Ephesian pro-
duction, but in other cases the mica is absent and the fabric
is more calcareous, with reddish and yellowish colouring
and some dark inclusions. The big formal variations in size
and fabric, found mainly in the M 273 corpus, make it dif-
ficult to achieve a clear comprehension and distinction of
all the different variants existing. Nevertheless some pieces
must be categorised within a subdivision of the M 273 and
described separately. This concerns the form identified for
the first time by Adamsheck from some upper parts found
in the Corinthian harbour of Kenchreai,168 and better docu-
mented some years later by Papadopoulos in the Chalcid-
ian locale of Torone from an almost complete piece.169 For
this reason the most appropriate name for this form, clearly
connected with the M 273, ought to be “Adamsheck 22/
Torone VII”. As already mentioned, an Icarian origin has
165
Bezeczky 2013a.
166
Degrassi et al. 2007.
167
We are not in agreement with those scholars who have tended to
extend the production of the Samos Cistern type till the second part
of the seventh century A.D. For us this form, first documented in a
mid-sixth-century context in the Heraion of Samos (Isler 1969), is
mainly found in sixth- and early seventh-century contexts and their
presence in some mid- and late seventh-century findspots could be
partially residual.
168
Adamsheck 1979.
169
Papadopoulos 1991; Papadopoulos 2001, 566-568, Fig. 155.
BEZECZKY_Text.indd 68
been suggested for this subtype typical of fifth century con-
texts in the eastern and central Mediterranean. We can only
confirm that within the Castrum material the small quan-
tity of these amphorae all have the typical calcareous, white
or yellowish fine fabric.
Unlike other places in the northern Adriatic and as in the
case of the relative frequencies of the LRA 1 and LRA 2, the
Samos Cistern group is present in the Castrum at a lower
level than their predecessors had been. Only a small quantity
of fragments of this late form has been documented at Cas-
trum. The fabric varies from one piece to another. Piece 101
has a fine and hard fabric, red at the core and light brown
outside, with some small red and colourless inclusions, simi-
lar to some Koan amphorae.
Even if the late Cretan amphorae did not make any impact
on the supply of the northern Adriatic, a couple of pieces
have been documented within the material of the Castrum.
One rim must be related to form TRC 1 or TRC 2 of Portale
and Romeo (57, 178 and 179). On the other hand, piece
180 looks to belong with the group named TRC 4, as pre-
sented by the two Italian scholars, and produced from the
fifth to early seventh centuries.170 At present no parallels of
this late form have been found in our region. The fabrics of
these two pieces are quite different, even if they both present
a regular quantity of small limestone inclusions.
After Cilicia and the Aegean, the Levant is the other main
oriental region that supplied amphorae to the Brijuni islands.
Unlike most places in the northern Adriatic, almost all the
amphorae documented from this region in this period are
Palestinian LRA 4; types such as Agora M 334 or LRA 5/6
have not been found. Equally the presence of LRA 4 within
our material is much more modest than in the main north-
ern Adriatic places, above all Ravenna, where they form the
most important amphora type.171 In Brijuni the number
of LRA 4 is less than the LRA 3, and so they comprise the
fourth most important oriental type of this period. Most of
the LRA 4 belong to forms of the sixth or early seventh cen-
turies, included in Peri’s group B; they present the typical
Gaza-Palestinian fabric, but some pieces are finer, more red-
dish and with fewer inclusions (99).172
In the first part of this article, the minor relevance of
other oriental production regions in the northern Adri-
atic was pointed out, particularly Egypt and the Black Sea,
whose products arrived just occasionally in this area. In Bri-
170
Portale – Romeo 2000, 422.
171
Cirelli 2007; Cirelli 2014.
172
Pieri 2005, 105-107.
18.03.19 14:39
 CHAPTER 6 – Eastern Mediterranean amphorae (González Cesteros)
juni, some fragments of Egyptian amphorae can be inserted
within the group of the LRA 7, namely the AE 7.1 group
of Dixneuf, dated mainly between the end of the fourth to
the mid-seventh centuries.173 Nevertheless it is possible that
piece 105 could be an early version of the group AE 7.2,
dated from the seventh to the end of the tenth centuries. It
is really difficult to identify both subtypes of LRA 7 from
just small fragments, since some parts of the amphorae, for
example handles and spikes, were probably handmade.174
The alluvial fabric of our pieces makes it easy to determine
the production place as within the Nile valley; they also
present several voids made by the disappearance or burning
out of organic components, mainly straw.
The Late Antique products of the Black Sea had even
less relevance than the Egyptian imports in our region, and
normally they are not found in excavations. In the case of
the Castrum there are some documented pieces that are
undoubtedly connected with the Sinopean production.
They belong to some amphorae of the so-called “Sinopean
Carrot amphorae” group, recently classified as lying within
the C-Sinope group as a result of the excavations of the
workshop of Demirci near Sinope.175 The Castrum pieces
belong to form C-Sinope II, with their different variants.
The fabric of these pieces is not always homogeneous, but
they do have some common features, mainly the use of black
volcanic stone fragments as temper, typical of the Sinopean
area. The presence of these fragments, even in an unimport-
ant quantity, must be stressed, since they are not found in the
north Adriatic contexts.
Even if it is not easy to associate a product with every
Late Antique oriental amphora with certainty, it seems quite
clear that most of the imports from the eastern part of the
Mediterranean that arrived at the Castrum were wines. Only
the LRA 2 are possibly linked with an olive oil content. No
oriental type found in Brijuni can be associated with fish
products. Nevertheless, some amphora forms, for example
the LRA 1 and the LRA 2, can transport different goods.
CONCLUSION
The staggering number of eastern Mediterranean amphorae
and the variety of forms present in the Brijuni Castrum seem
to be in concordance with the general pattern of the north-
173
Dixneuf 2011, 154-173.
174
Dixneuf 2011, 155-156.
175
Kassab Tezgör 2010.
BEZECZKY_Text.indd 69
69
ern Adriatic, a region that maintains a special link with the
eastern regions by the maritime routes of the Aegean.
Nonetheless, some peculiarities of this material draw
attention to themselves: probably they exist thanks to the
special character of the site, situated in the estates of a very
rich family that would later become Imperial property during
the early and middle Roman period, and form a possible mili-
tary and ecclesiastical base during Late Antiquity. Its nearest
most important Istrian city is Pula, only around 10 kilometres
from the island. The shipping of the amphorae supply from
this important harbour, as well as from Fažana, is the most
logical option. Nevertheless the singularity of Brijuni seem-
ingly ensures the occurrence in the amphorae spectrum both
of other forms and also of small variations in the respective
percentages present, compared to other places in the region,
such as, for example, Aquileia or Trieste. A complete publica-
tion of the material in Pula could help solve several questions,
as well as providing a deeper appreciation of the social and
economic development of the whole Istrian Peninsula.
During early Imperial times the amphora imports from
regions outside the Adriatic have no perceivable relevance at
Brijuni. The existence of several presses on the island demon-
strates the production of wine in this period and, we are sure,
of olive oil too. The large number of Dressel 6B documented
must be assigned to this foodstuff production. Within this
trend, the oriental amphorae of early Imperial times are the
most represented imports coming from outside the Adriatic
region. This pattern looks to conform to the norm as seen
in other places in the northern Adriatic. However, there is
one main difference with places such as Aquileia, Trieste or
the most important Cisalpine cities: at Brijuni, the number
of Rhodian amphorae is outnumbered or at much the same
level as the Koan and Knidian imports.
The abrupt end of Istrian amphora exports during the
Antonine period had an extraordinary impact on life at Bri-
juni: much less material was brought in now, due to the end
of the Dressel 6B production and their gradual substitution
by other types, mainly Fažana 1 and Fažana 2, that nonethe-
less never reached the massive volume of their predecessors.
Even if it is true that deposits of the late second and third
century layers in the Castrum are not so well represented as
others, nonetheless the number of amphorae assignable to
this period decreases proportionately. The oriental forms
still count as one of the most represented overseas import
groups: they make up a diverse group of eastern Mediter-
ranean amphorae from different regions conveying a varied
range of but one product. All the types documented are wine
amphorae; forms linked with olive oil, like the Dressel 24
18.03.19 14:39
 70 THE AMPHORAE
or the MRA 8, commonly found in some of the best exca-
vated places in the northern Adriatic, do not appear at Bri-
juni. This fact, together with the arrival and maintenance of
new Istrian amphora types that could be linked with olive
oil, could indicate the limited revival of the Istrian olive oil
production, some years after the abrupt interruption that
probably took place during Hadrian’s reign.176
Within the oriental amphorae, most of the forms came
from the Aegean basin, but from the end of the first century
A.D. forms appear from other regions such as Crete or Cili-
cia. Nevertheless, the most represented types are the several
versions of “one-handled jar”, a typical production from the
Cayster valley and Ephesus region, but also produced in
other workshops throughout the Asia Minor territory and
nearby islands.
The Late Antique material shows a completely differ-
ent world-view for the Brijuni islands. The administrative,
military and religious changes that took place in the whole
Roman world from the late third or early fourth centuries
A.D. also have their effect on the architectonical and material
remains of the so-called Castrum. Even if it is possible that the
Castrum still produced wine during this period and maybe
olive oil too, as the existence of some presses from this period
surely argues,177 it is undeniable that its function has been
transformed from a productive villa rustica to the quarters of
a military unit. Some hundred meters from the Castrum, a
church is destined to be built in the fifth century or even earli-
er.178 It is in Late Antiquity that the oriental amphora imports
are at their zenith. Even if this chronological phase is prob-
ably overrepresented, because of the nature of the settlement
at that time, nonetheless the volume of imports from outside
the Adriatic is now much higher than earlier on.
176
Bezeczky 1998a, 10-11.
177
Matijašić 1993a, 249-251.
178
See Schobert in this volume.
BEZECZKY_Text.indd 70
Several oriental regions exported their amphorae prod-
ucts as far as Brijuni during this period, with Cilicia, the
Aegean and the Levant being the most important among
them. The huge quantities of LRA 1 overwhelm all other
oriental imports into second place. This dominance repre-
sents the big difference compared with other places in the
region, but there are some smaller particularities in the per-
centages too. The LRA 2 are really well-represented within
the material and the Ephesian products too. The presence
of some Ephesus 56, an almost non-existent form in the
northern Adriatic, is exceptional and constitutes a central
difference from any other context in the region, at least up
until now. Other products, such as late Cretan and Black
Sea amphorae types present in Brijuni, are also unexpected
in the region, but their number is really insignificant in the
Castrum. Lastly, the Levantine products, mainly the LRA 4,
are found in Brijuni too, but they are not worthy of compar-
ison with numbers in other places in the northern Adriatic,
especially with Ravenna, which seems to have been their
main market.
The presence of one fragment of the neck of a probable
LRA 13 with a Byzantine Imperial stamp and dated to the
very late sixth century A.D. is the latest oriental form docu-
mented with certainty in the Castrum.
The amphora study of Brijuni contributes to the eco-
nomic and social history of Istria; other studies will hope-
fully continue, taking up the baton. We see Brijuni, and Istria
generally, not only as an producer of amphorae and olive oil
production to a major degree during the Roman period, but
also as a lively market on a most important trade route in the
Adriatic – one open to the whole Mediterranean world, with
especially close ties to the eastern part.
18.03.19 14:39

7 AFRIC AN A MPHOR AE
Michel Bonifay – Claudio Capelli
In this chapter, we are dealing with containers attributed to
Roman Africa, namely the Roman provinces of Mauretania
Caesariensis, Numidia and Africa Proconsularis, gathered
from the beginning of the fourth century onwards into the
dioceses of Africa (at the same time as the new provinces of
Sitifensis, Byzacena and Tripolitania were created), and cor-
responding to the modern-day (coastal) territories of Algeria
and Tunisia, and of the western part of Libya.
Sixty-one sherds of African amphorae have been studied,
belonging to thirty different types, and dated from the mid-
second century to the seventh century A.D. An accurate
graphic and photographic documentation together with
a visual examination of each sherd followed by a binocular
examination of all amphora micro-samples and by a petro-
graphic analysis of a series of selected thin sections allowed
us to specify the typology and the origin of most of these
containers.1 Associated to the chronology and the content
hypothesis currently suggested for each of these types, this
amphora study sheds new light on African foodstuff dis-
tribution during the Roman and Late Antique times in the
northern Adriatic.
1 TYPOLOGY AND ORIGIN OF THE AFRICAN
AMPHORAE
1.1 AFRICAN AMPHORAE OF PUNIC
TRADITION
A single sherd can be attributed to this family. The foot 108
is very characteristic of type Hammamet 2E or 3A of the late
fifth-start of the sixth century.2 Similar palm marks incised
1
The whole graphic and photographic documentation is due to Tamás
Bezeczky, our thanks to him. African amphora sherds were extensively
examined and discussed in Brijuni by Bezeczky and Michel Bonifay
in June 2013. The petrographic study of the micro-samples and the
thin sections of African amphorae was performed by Claudio Capelli
in Genova in 2014 and interpreted in collaboration with Bonifay in
Aix-en-Provence in 2015. In the catalogue, the thin section descrip-
tions are to the work of Capelli. The paper was written in 2015.
2
Bonifay 2004, Figs. 50.1 and 51.3.
BEZECZKY_Text.indd 71
71
before firing are already known on the same amphora type in
Tunisia.3 The fabric fits very well with an origin in the north-
ern Hammamet gulf, probably in the region of Sidi Jdidi,
where a huge workshop has recently been evidenced.4
1.2 ROMAN-AFRICAN AMPHORAE
1.2.1 CLASSICAL ROMAN-AFRICAN
AMPHORAE
No Tripolitanian amphorae were found, except a single dubi-
ous sherd (109), similar in shape to type Tripolitana III (?),
but the fabric is not typical for a Tripolitanian origin.
A first, small group includes classical African amphorae
of types Africana I and II of the mid-second to mid-third
centuries. Among these the Sullecthum/Salakta products
are well attested, with sherds of Africana I A (110) and II
A (112). The attribution to this workshop is very easy, even
with the naked eye, due to the very distinctive, brick-red
and/or grey fabric, rich in small white inclusions (lime and
micro-fossils inclusions, alongside the usual rounded quartz
grains)5; types Africana I A and II A were star products dur-
ing the second-third centuries in Sullecthum, as demon-
strated by the most recent workshop surveys.6 Other pro-
duction regions are not completely absent from the Brijuni
assemblage. No. 111, provisionally identified as a variant of
Africana I (?), could originate from a northwestern zone of
Africa Proconsularis, perhaps the same zone where the type
Ostia XXIII was produced,7 while no. 113 is typologically
very similar to the Central-Tunisian production of Africana
II ‘pseudo-Tripolitanian’.8
3
Wadi Arremel: Bonifay 2004, Fig. 8.17. Sidi Jdidi: Mukai 2010,
Fig. 170, context 169.2. From now see Mukai 2016.
4
Mukai 2014: workshop of Tefernine.
5
Capelli et al. 2006.
6
Nacef 2015a.
7
Bonifay et al. 2015, 189-210. Hypothesis advanced more from a
typological rather than a petrographical point of view.
8
Bonifay 2004, 114 and Fig. 110, type 24. In this case, petrography
cannot help.
18.03.19 14:39
 72 THE AMPHORAE
A second group, later in date, brings together Africana II
types of the mid/second half of the third century. The
Nabeul origin of type Africana II C (114 and 115) is con-
firmed by the petrographical macroscopic observation. This
geographical attribution permits us to classify the foot 115
as Africana II C instead of Africana II D, because this first
type is very characteristic of the Nabeul amphora produc-
tion.9 No. 114 could be a late variant C3 of the same type,
not predating the beginning of the fourth century. Type
Africana II D was produced in several cities of the central
Byzacena coastline, at least in Hadrumetum, Leptiminus
and Salakta, but no. 116, according to the petrography, more
likely originates from Thaenae, where the production of this
type has also been evidenced.10 Finally, the origin of base 118
and rim/neck 119 cannot be determined.11
The third, most furnished group is devoted to type Afri-
cana III or Keay 25, from the fourth century. A first sherd
(117), transitional between type Africana II D and type
Africana III A, remains of unknown African origin. Within
the variant Africana III A (= Keay 25.1), two workshops
already encountered, Thaenae (?) (120) and Nabeul (122
and 123), have been identified while no. 121 remains unde-
termined. When the sherds are small, it is difficult to use
the typology because type Africana III A was produced
almost everywhere in Africa, and we have to deal with very
discrete morphologic variations. By contrast, the variant
Africana III B (= Keay 25.3) is considered as being typical
of Nabeul.12 Indeed, among the three sherds attributed to
this type, two (124 and 125) seem to have been produced
in Nabeul, while the last one (126) does not. Lastly, a sin-
gle sherd (127) can be identified as Africana III C (= Keay
25.2), originating from Nabeul, according to the petrogra-
phy and the typology.
1.2.2 LATE ROMAN-AFRICAN AMPHORAE
SMALL-SIZED LATE ROMAN-AFRICAN
AMPHORAE
Nine sherds have been attributed to the so-called ‘spatheion’
type13 of the fifth century. The first one (128), dubiously clas-
sified as spatheion 1, and originating from Nabeul, could also
9
Bonifay 2004, 115.
10
Bonifay 2004, 33.
11
In this case, the typology, which is not clear, could not help.
12
Bonifay 2004, 122.
13
On this erroneous but universally admitted denomination, see Boni-
fay 2004, 125.
BEZECZKY_Text.indd 72
be identified as a large-sized Keay 35A amphora. Sherds 129
to 131 are more characteristic of type spatheion 1. According
to the typology and the petrography, two of them (130 and
131) come from Nabeul, the latter probably being later in
date.
Traditionally included in the ‘spatheion’ family, even
if more probably completely distinct in shape and in date,
the so-called ‘miniature spatheia’ (type 3), dating back to
the end of the sixth and the seventh century, are well rep-
resented in Brijuni. A first sherd (132) is difficult to clas-
sify within the current typologies, while its fabric does not
fit with any known references, except some Algerian ones.
No. 133 could be identified as a variant B of type spatheion
3, with a petrography very similar to the amphora produc-
tion of northwestern Zeugitana (namely between Utica and
Tabarka). Lastly, a Nabeul origin is confirmed for the three
examples of spatheion 3C (134-136), the latter being the less
characteristic from a petrographical point of view, for such
an allocation of origin.
MEDIUM-SIZED LATE ROMAN-AFRICAN
AMPHORAE
A new type of late Roman-African amphora is currently
being identified from rim/neck fragments scattered around
the western part of the Mediterranean,14 characterised by fre-
quent epigraphic marks vertically incised on the neck before
firing.15 One of these finds comes from the Sidi Zahruni
workshop in Nabeul,16 the city where, it seems, according to
the petrography, that almost all these amphorae where pro-
duced. During the current investigation on this type, some
complete exemplars have been identified,17 none of them
being published until now.18 This amphora type is about
85 cm high for a body diameter of 26 cm, so looks like a
medium-sized amphora. We provisionally propose identify-
ing this new type with the fragmentary rim/neck published
by Simon Keay under his type 3/5.19 One sherd (137) might
belong to this type, with a classic Nabeul-zone B fabric.
14
Capelli et al. 2016, 288.
15
Portus: Franco 2012, Fig. 5.8. Aquileia: Cipriano – Carre 1987,
Fig. 19; Bueno et al. 2012, Fig. 6.
16
Bonifay 2004, Fig. 20.5.
17
For example, in the Mercati Traianei collection at Rome (inv. no.
ANF 347, 348, 350) and in the storeroom of the Aquileia museum.
18
See, however, a possible comparison with a complete exemplar found
in Algeria and published without scale: Salama 2006, Fig. 23.
19
Keay 1984, Fig. 42.4.
18.03.19 14:39
 CHAPTER 7 – African amphorae (Bonifay – Capelli) 73

LARGE-SIZED LATE ROMAN-AFRICAN appear to have been produced in the Sahel region of Tuni-
AMPHORAE sia. The Moknine workshops are represented by examples
of Keay 62 (153 and 154) and Keay 61 A/D (158-159)
The large-sized late Roman-African amphorae are the most
amphorae. The latter are characteristic of the ‘classic’ fabric,
numerous. Two first sherds (138 and 139) belong to type
evidenced in the workshop dump excavated at the beginning
Keay 59/8B, with a very distinctive pink, compact fabric,
of the 90s,26 while the first do not seem to fit with the new
and beige, smooth surface. The rim 138 illustrates a vari-
workshop dump recently discovered by Jihen Nacef in the
ant of type Keay 8B dating back to the second half of the
same city.27 Another sherd (161) must be classified as type
fifth century,20 while the foot 139 could belong to the one
Keay 61C, and subsequently attributable, from both a typo-
or the other of the two sub-types Keay 59 or 8B. The origin
logical and petrographical point of view, to the workshop
of this Keay 59/8B type has been attributed to south Byz-
of Henchir Chekaf in the hinterland of Salakta.28 The type
acena, based on the discovery of two workshops.21 Another
Keay 8A is also present (162-163) and its origin in the Sahel
sherd (140) belongs to a barely documented type, classified
region seems probable, even if it is not possible to specify
by S. Keay as a variant of his type 11B, even if the rim profile
the workshop itself. The production of this type is evidenced
is very different from the main (Tripolitanian) type.22 The
in Leptiminus29 and near Thapsus.30 Lastly, a southern Sahel
petrography cannot help us to localise the production, but
provenience (the region of Rougga?)31 has been confirmed
some similarities are observed with the fabrics of the previ-
for the three sherds of type Keay 34 (164-166) discovered
ous Keay 59/8B type. The presence of both these types in the
in Brijuni.
survey of the Kasserine region23 could be an argument for a
common, south Byzacena origin. A fourth sherd (143) is a
Keay 36 amphora, with a fabric entirely compatible with the 1.3 AFRICAN IMITATIONS OF NON-AFRICAN
normal fabric of the Keay 27 and 36 amphorae. According TYPES
to the distribution of these types in Africa, an origin from
A single sherd (167) could be attributed to a variant of type
the Carthage region or the southern Mejerda valley seems
Keay 1B, which is a fourth-century late derivative of type
likely.24
Gauloise 4, perhaps produced in the territory of modern-day
Then, the Brijuni assemblage provides a large series of
Algeria. However the fabric of this sherd is quite different
large-sized amphorae originating from Nabeul, including
from the normal fabric of this type.32
types Keay 35A (141), 35B (142), 55 (144 and 145), 57
(146-148), 55-56-57 (foot 149), from the fifth century-first
half of the sixth century. Most of these amphorae, easy to 1.4 STORAGE JARS (?)
classify within Keay’s typology, appear to come, according
Two fragments of massive feet (168 and 169) do not seem
to the petrography, from zone B of the Nabeul workshops,
to belong, because of their size, to ‘normal’ transport ampho-
active between the fifth and the seventh centuries.25 On the
rae. They may fit better with African storage amphorae, well
other hand, as elsewhere, type Albenga 11/12 or Keay 62Q
known in the African territory,33 and occasionally travelling,
amphorae (150-152), from the second half of the fifth cen-
as evidenced by examples found in Rome34 and even in a
tury-first third of the sixth century, are difficult to classify,
shipwreck along the Croatian coastline.35
and their origin(s) remain unknown until now.
By contrast, the latest productions from the the mid-sixth
to the end of the seventh century are easier to characterise.
Except for two feet of type Keay 62 or 61 (155 and 160),
the first one perhaps attributable to the Nabeul workshop
and the second one of unknown origin, all these amphorae 26
Bonifay 2004, 35; Gandolfi et al. 2010, 37-38.
27
Nacef 2015b.
28
Bonifay 2004, 140; Nacef 2007; Gandolfi et al. 2010, 37.
29
Dore 2001, Fig. 1.65, no. 18.
20
See, for example, Pieri 1998, Fig. 198.280-282. 30
Nacef 2014, Fig. 3.26-33.
21
Bonifay 2004, 132. 31
Bonifay 2004, 143.
22
Keay 1984, Fig. 172.2. 32
Capelli – Bonifay 2007, 554-555.
23
Neuru 1990, Figs. 16.j and k. 33
Bonifay 2013, 539 and Fig. 1.
24
Bonifay et al. 2012, 241. 34
Contino – D’Alessandro 2014, 328 and Figs. 9-12.
25
Bonifay et al. 2010. 35
Huguet 2012, 130 and Fig. 50.2.

BEZECZKY_Text.indd 73 18.03.19 14:39

 74 THE AMPHORAE
2 AFRICAN IMPORTS TO BRIJUNI IN THE
NORTHERN ADRIATIC CONTEXT
A first, previous survey on African imports to Brijuni had
demonstrated that African amphorae represented about
40 % of the whole late Roman amphorae, while they were
completely absent from the early Roman contexts.36 In con-
sequence, if we want to gauge the imports of African ampho-
rae to Brijuni, it is important to periodise these imports, and
to look for comparisons with other northern Adriatic sites as
well as other African ceramic imports. The specific history of
the Roman villa and the Byzantine ‘castrum’ could also help
with the interpretation of this documentation.
BEFORE THE MID-SECOND CENTURY
Although the villa existed from the first century B.C.,37 no
African amphorae found in Brijuni seem to predate the mid-
second century A.D. Almost the same situation prevails
in the underwater finds along the Croatian coastline.38 Of
course, the percentages of African amphorae are low every-
where during the late Republican period and the beginning
of the Empire, even in Rome where the African (wine?) con-
tainers generally represent less than 10 % of the total amount
of the amphorae in the mid-first-century A.D. contexts.39
But on the other hand, the African imports of oil, olives and
wine become well attested during the time of Flavian not
only in Rome, but also in Pompeii, and in Arles, with the
types Tripolitana I, Ostia LIX and Schöne-Mau XXXV. It
is likely that the gap in early African amphorae will soon be
filled, at least in part, by new research in the northern Adri-
atic. Some Flavian types are already attested in Zaton40 and
in Aquileia,41 suggesting some imports of African wine and
salted fish (?). Probably, the inhabitants of the ‘Castrum’
villa of Brijuni did not need to consume these imported
foodstuffs, although they already felt the need to use some
African table- and cooking wares.42
36
Vidrih Perko – Pavletić 2000, Figs. 2 and 3.
37
Begović – Schrunk 2011a, 376.
38
Pešić 2013, 1210.
39
Reynolds 2010, table 1a.
40
Pseudo-Dressel 2/4 (wine) from western Tripolitania. Personal obs. in
2013, thanks to Dr. S. Glušcević.
41
At least three complete exemplars of type Uzita 52.10 (salted fish) in
the collections of the Museo Archeologico Nazionale. Personal obs. in
2013, thanks to Dr. P. Ventura, Director of the Museum.
42
Vidrih Perko – Pavletić 2000, 264 (ARS forms Hayes 8A and
9A), and 265 (early ACW form Hayes 198).
BEZECZKY_Text.indd 74
MID-SECOND CENTURY TO END OF THE
THIRD CENTURY
As early as the mid-second century, the new type Africana I
reached the northern Adriatic in such quantities that they
could be reused for the packaging of local salted fish, as
shown by the Grado shipwreck.43 Following the drop in the
production of Histrian olive oil, and in spite of the massive
import of Spanish oil containers Dressel 20, the need for
olive oil, not only for local use but also for redistribution
towards the north, ensured an outlet for the African produc-
tion surplus. As in many places in the Mediterranean and
also within the Grado shipwreck cargo, the Salakta products
are the first ones to be attested in Brijuni (110). The same
port of Byzacena also provided other foodstuffs, packaged
in Africana II A amphorae, possibly salted fish or even wine
(112). The presence of the Sullecthum amphorae along the
eastern Adriatic coastline, associated with central Tunisian
ARS A/D44 and C, has sometimes been interpreted as evi-
dence of a direct connection between Byzacena and the
Adriatic, but the petrographical observations on the late
Antonine and Severan amphorae from Brijuni show that
some other African regions (111?, 113) were exporting their
olive oil surpluses as well.
The situation seems to change from the mid-third cen-
tury onwards, with the emergence in the Brijuni assemblage
of new exporting cities, like Nabeul and Thaenae, and with
the supremacy of other foodstuffs rather than olive oil: prob-
ably salted fish and/or wine packaged in Africana II C (115)
and II D amphorae (116). These types are well attested in
Aquileia.45 But it does not mean that African olive oil did
not reach the Adriatic anymore. Africana I B amphorae are
present in Aquileia46 alongside Tripolitana III amphorae,47
including some late variants,48 evidence of continued imports
of oil from the Severan period onwards.
43
Auriemma 2000.
44
Some sherds have been found in the ‘Castrum’ villa: Pröttel 1996,
Pl. 47. 12-13.
45
Africana II C: Verzár Bass 1994, Pl. 66, AA 15. Africana II D:
Verzár Bass 1991, Pl. 38, AA 29; Cipriano – Carre 1987,
Fig. 20.
46
Carre et al. 2007, 608 and note 7.
47
An example of Tripolitana III with a stamp AVGGG in Aquileia:
Cipriano – Carre 1987, Fig. 18.
48
Verzár Bass 1994, Pl. 66, AA 2.
18.03.19 14:39
 CHAPTER 7 – African amphorae (Bonifay – Capelli)
FOURTH CENTURY TO BEGINNING OF THE
FIFTH CENTURY
Abundant in the Brijuni assemblage, the fourth-century
Africana III amphorae are also the most widespread African
containers throughout the Roman world. These amphorae
generated the highest percentages of African imports in
Rome,49 and provided impressive quantities of exemplars in
the northern Adriatic50, often reused in drainage systems.51
The hypothesis of a wine content, put forward for these
amphorae according to their typology, their systematic
internal pitch lining, and the ‘universal’ pattern of their dis-
tribution, has recently been confirmed by a series of chemi-
cal analyses,52 even if another content could be possible for
variant B. The ‘Castrum’ villa – or already ‘settlement’ – of
the fourth century imported a good deal of African wine (at
least 120-123, and possibly also 124-127).
Africana III amphorae also are “the most common find
in the underwater sites in Croatia”.53 Africana III wrecks are
particularly abundant on the eastern side of the Adriatic,
which is considered the ‘upward’ direction of sailing. One of
the most important cargoes is illustrated by the Sobra wreck,
from the middle-third quarter of the fourth century, with
an assemblage of Africana III A and B, some of them bear-
ing a stamp PGT attributable to the city of Nabeul.54 Two
examples of the same stamp have been found in Aquileia.55
These observations are consistent with the Brijuni assem-
blage where most of the Africana III amphorae appear to
originate from Nabeul.
The comparison between the Brijuni and Aquileia assem-
blages extends to some rare types, such as a variant of Keay
1B (167) attested at both of these sites.56
At the same time, Aquileia and the ‘Castrum’ villa/settle-
ment were still being supplied with African table- and cook-
ing wares. In Brijuni, high-quality C4 ARS Hayes 53A with
applied decoration are present, alongside cruder D1 plates
Hayes 59 and 61A with stamped decoration of styles A(i)
and (ii).57
49
C. Panella in Anselmino et al. 1986, Table 3; Panella – Saguì
2001, 766-767.
50
Carre et al. 2007, Fig. 4.
51
See, for example, Maurina 1999.
52
Woodworth et al. 2015, 41-57.
53
Pešić 2013, 1208.
54
According to recent petrographical analyses.
55
Verzár Bass 1994, Pl. 76, AB 7-8.
56
Verzár Bass 1991, Pl. 38, AA 42; Verzár Bass 1994, Pl. 69,
AA46.
57
Pröttel 1996, Pl. 48, 54.6-11, 85-87, 89-91.
BEZECZKY_Text.indd 75
75
FIRST HALF OF THE FIFTH CENTURY TO FIRST
HALF OF THE SIXTH CENTURY
The period comprised between the first half of the fifth cen-
tury and the first half of the sixth century is a time of massive
inflow of African amphorae into the fortified settlement of
Brijuni. Paul Reynolds already noted this trend in the whole
Adriatic from the mid-fifth century onwards, explaining
it by the ‘major shipping route to Ravenna’, and the ability
of cities and settlements located along this route to benefit
from these exports.58
A lot of different types are present in Brijuni, already well-
known in different places on the Mediterranean, sharing in
common a strong presence of Nabeul products: spatheia
type 1, Keay 3/5, 35A and B, 55, and 57 (128-131, 137,
141-142, 144-145, and 146-148), for example in Catalo-
nia where they were first classified,59 in southern Gaul,60 in
Rome,61 and even on the Black Sea.62 In some places, such as
in Brijuni, they are associated with the south Byzacena types
Keay 59/8B (138-139), northwestern types Keay 27/36
(143), and the unprovenanced type Keay 62Q (150-152).
All these amphorae are well attested along the eastern side of
the Adriatic, in Albania,63 Croatia and the Slovenian Karst
region64, as well as in Aquileia,65 Ravenna66 and along the Po
Valley.67
A major problem posed by this documentation is that
we know little about the content of these containers. This
uncertainty is further increased by the fact that the spatheion
type was possibly a cargo space-filler with an interchangeable
content of oil, fish sauce or wine. Reuses of these amphorae
are also possible, as demonstrated by the case of the ware-
house of Classe, which burned down at the end of the fifth
century.68
In Brijuni, these amphorae are accompanied by huge
amounts of African tableware, with a pattern very similar to
the one known in the northern Gulf of Hammamet and in
58
Reynolds 2010, 99. See also Auriemma – Quiri 2007, 34.
59
Keay 1984.
60
Marseille: Bonifay et al. 2011.
61
Panella et al. 2010, 66-67.
62
Opaiţ 1997-98.
63
Butrint: Reynolds 2004, Fig. 13.155. Byllis: Bonifay – Cerova
2008. Durres: Hoti – Capelli – Piazza 2008; Shkodra 2005,
Fig. 10.2, 4-5. Shkodra: Hoxha 2003.
64
Vidrih Perko 1997.
65
Collection of the Museo Archeologico Nazionale, personal obs. in
2013.
66
Antiquarium of Classe, personal obs. in 2006.
67
Brescia: Brogiolo 1999. Milan: Lusuardi Sienna 1990.
68
Pecci et al. 2010.
18.03.19 14:39
 76 THE AMPHORAE
Nabeul: forms Hayes 50B.61, 61C, 87A, Fulford 27,69 sug-
gesting a direct connection between the Cape Bon region
of Tunisia and the northern Adriatic for both the amphorae
and the tableware. Another interesting feature of the fifth-
century African imports in the northern Adriatic is the pres-
ence of huge quantities70 of flanged bowls, Fulford 22-23,
again produced in Nabeul.71
MID-SIXTH CENTURY TO LATE SEVENTH
CENTURY
The last period of African imports to Brijuni mainly inclu-
des products from coastal Central Byzacena: Keay 62 and 61
from Moknine, Keay 61C from the Salakta hinterland, and
Keay 8A from unknown Tunisian Sahel sources (153-154,
156-159, 161, and 162-163), alongside miniature spatheia
of type 3C from Nabeul (134-136). This kind of assemblage
has sometimes been interpreted as a signature of the Praeto-
rian Prefecture organisation72 and of the specific supply of
the Byzantine territories. At this time, the supply of Brijuni
is identical to the supply of the Byzantine place of Koper,73
69
Pröttel 1996, Pl. 52.21, 52.8-9, 50.14, 47.8.
70
Brijuni: Pröttel 1996, Pl. 52.17. Aquileia: Verzár Bass 1994,
Pl. 45-46.
71
Bonifay 2004, 255 and Fig. 139.
72
Reynolds 2010, 127.
73
Cunja 1996.
BEZECZKY_Text.indd 76
including some tableware, ARS forms Hayes 105, 106 and
109.74 Does this character imply that, at this time, the Bri-
juni ‘Castrum’ had become a Byzantine military settlement
as well?75
No precisions about the contents of these amphorae can
be obtained from the current state of the research.
CONCLUSION
Even if this documentation is no longer contextualised,
a careful integrated archaeological and petrographical
approach to the individual sherds leads to a better under-
standing of the supply of the Brijuni Islands with African
foodstuffs from the mid-second century to the end of the
seventh century. This assemblage represents an important
milestone for the study of the African amphorae between
the sites recently surveyed in Albania (Butrint, Byllis, Shko-
dra), and the main cities of the northern Adriatic (Aquileia,
Ravenna), and probably the first one to have been extensively
studied with the specific aim of specifying the origins of the
amphorae within the African territory.
74
Pröttel 1996, Pl. 52.1-2, 4-6.
75
Begović – Schrunk 2011a, 384.
18.03.19 14:39

8 A MPHOR A STA MPS AND INSCRIPTIONS
Piero Berni Millet – Tamás Bezeczky
The catalogue contains the inscriptions and stamps on the
amphorae found at the villa. However, we wanted the reader
to be able to study them all together, as in the 16-18 tables.
2. Lamboglia 2, rim, neck, no. 533 (152) – 3960, Pl. 1 and 16.
Stamp: BAR·ME[---], on the rim. This stamp has not been
found elsewhere.
P. Desy mentions the BARNAES and G. Volpe another
BARNA (retro) stamps.1 The Tarentum [---]BARN stamp
is probably fragmentary.2 However, the Delos BARN stamp
could be connected to the Castrum stamp.3
3. Lamboglia 2, rim, no. 656 (118), Pl. 1 and 16.
Stamp: VERSO, on the rim.
The other Castrum stamp, VERSO, is widely known.4
Brunella Bruno distinguished two VERSO stamp types, but
the Castrum version is somewhat different from both.5 In
Vercelli dated in second part of the first century B.C.6
5. Dressel 6A, rim, neck, no. 140 (54) – 25973, Pl. 1 and 16.
Stamp: V, V(…), on the rim. This stamp has not been found
elsewhere.
7. Dressel 6A, rim, neck, no. 140 (153) – 25972, Pl. 1 and 16.
Numeral graffito on the rim, cut after firing: XXIIII, perhaps
‘twenty-four’.
1
Desy 1989, 179; Volpe 1990, 245, Fig 241, no. 3 = Blanc-Bijon
et al. 1998, 55, no. 610.
2
Desy 1989, 26, no. 46.
3
Desy 1989, 59, no. 327; CIL III. 7309,35.
4
Desy 1989, nos. 116, 118, 281, 299, 302, 358; Bruno 1995, 110-
112; Carre et al. 1995, 47, no. 96 bis; Gabucci – Quiri 2008, 65;
Italy (Milano, Ivrea, Altino, Luni), Albania (Durrhachium, Kryemd-
hej), Greece (Delos), Spain (Cabezo Augudo, San Jordi A, Villaricos,
Barcelona).
5
Bruno 1995, nos. 110-111.
6
Brecciaroli Taborelli 1987, 118.
BEZECZKY_Text.indd 77
77
Laecanius (nos. 11-17) and Emperor’s (nos. 18-22) produc-
tion, in Fažana workshop.
Stamp - Dominus stamp: the Laecanius or Emperor’s stamp
Stamp* - vilicus or workshop manager stamp
11. Dressel 6B, rim fragment, no. 1493 (01), Pl. 2 and 16.
Stamp: [LA]EK·A, <C.> [La]ek(ani) <Bassi> A(...).7
Occurrence: Magdalensberg, Mantua, Novaria, Este, and
Verona.
12. Dressel 6B, rim, neck fragment, handle, no. 38 (25) –
2129, Pl. 2 and 16.
Stamp*: [S]YNT, [S]ynt(…).8
Occurrence: Dertona, Muntia, Patavium (2), Vercellae, Vin-
centia (2).
13. Dressel 6B, rim fragment, no. 1326 (611), Pl. 2 and 16.
Stamp*: COMVS, Comus, S retro.9
Occurrence: Magdalensberg.
14. Dressel 6B, rim, no. 1243 (110), Pl. 2 and 16.
Stamp: LAE, <C.> Lae(kani) <Bassi>.
15. Dressel 6B, rim, no. 719 (117), Pl. 2 and 16.
Stamp: C·L[AEK], C. L[aek](ani) <Bassi>.
Occurence: Pula, Celeia, Poetovio.10
16. Dressel 6B, rim, no. 1279 (123), Pl. 2 and 16.
Stamp: [LA]EK, <C.> [La]ek(ani) <Bassi>.
17. Dressel 6B, rim, no. 1371 (121), Pl. 2 and 16.
Stamp*: PA[GANI], Pa[gani].
7
Bezeczky 1998a, 95, no. 1; Mazzocchin 2013, 121, nos. 25.1a-d.
8
Perhaps Syntrophus and as Syntychus, Mócsy et al. 1983, 279; Pape
1884, 1459; Bezeczky 1998a, 204, no. 507; Mazzocchin 2013,
nos. 25.14a-b.
9
Mócsy et al. 1983, 86; Bezeczky 1998a, 232, no. 611. Pl. 32 and 56.
10
Bezeczky 1998a, nos. 37-38.
18.03.19 14:39
 78 THE AMPHORAE
Occurence: Pullaria ins.(Verige), Aguntum (2), Iuvavum,
Augusta Taurinorum, Concordia, Dertona, Opitergium (2),
Patavium, Cervignano.11
18. Dressel 6B, rim fragment, no. 1476 (648), Pl. 2 and 16.
Stamp: IM[P], Im[p](eratoris) [Vespasiani].12
Occurence: Fažana, Pullaria ins., Dragonera, Aguntum,
Aquincum, Concordia, Flavia Solva, Tokod, Augusta Tauri-
norum.
19. Dressel 6B, rim, no. 1338 (120), Pl. 2 and 16.
Stamp: IMP·TRA, Imp(eratoris) Tra(iani), upside down
Stamp*: ...VS.13
Dominus stamp occurence: Mediolanum, Vindobona,
Savaria.
20. Dressel 6B, rim, no. 1426 (122), Pl. 2 and 17.
Stamp: [IM]P·TRA, [Im]p(eratoris) Tra(iani), upside down.
Occurence: Mediolanum, Vindobona, Savaria.
21. Dressel 6B, rim, no. 12 (116) – 26017, Pl. 2 and 17.
Stamp: [I]MP·T[RA], [I]mp(eratoris) T[ra](iani).
Occurence: Mediolanum, Vindobona, Savaria.
22. Dressel 6B, rim, no. 500 (119) – A 10627, Pl. 2 and 17.
Stamp*: LESBI (reverse).
Occurence: Concordia, Patavium (2), Siscia, Savaria,
Augusta Taurinorum.14
38. Beltran IIA, base, no. 10 (6) – 16027, Pl. 17.
Numeral graffito cut after firing: V, perhaps ‘five’.
40. Dressel 20, handle, no. 1271 (125), Pl. 17.
Stamp: palm branch (ramus palmae) M[---].
Guadalquivir production, similar stamp (ramus palmae)
MAF (ramus palmae)15
The palm branch was in vogue during Hadrian’s period.
Occurence: Augst.
41. Dressel 20, body, base, no. 1554 (115), Pl. 5 and Ch. 5
Figs. 10, 12 and 13.
Graffito cut before firing: Asiatici / sec( ) × k(alendas)
mar(tias) Alb( ) + II.
Date: Second third of the second century A.D.
11
Bezeczky 1998a, nos. 490, 652-659; Cipriano – Ferrarini
2001, 167-168; Bezeczky 2014, 250.
12
Bezeczky 1998a, 240, no. 648. Pl. 34 and 56.
13
Bezeczky 2014, 251.
14
Bezeczky 2014, 251.
15
Martin-Kilcher 1987, 94, (1388), ST 2a.
BEZECZKY_Text.indd 78
Similar graffito occurence: Germany: Altenstadt (Hessen),
England: Lincoln (Lincolnshire), Wall (Staffordshire),
Brough on Noe (Derbyshire), Vindolanda (Bardon Mill),
France: Saint-Thibéry (Hérault), Béziers (Hérault), Reims
(Marne).
See Berni Millet in this volume.
86. Late Roman Amphora 1 (LR 1), rim, neck, handles, no.
1342 (136), Pl. 17.
Graffito cut after firing: A( ).
100. Late Roman Amphora 13 (LR 13), neck, no. 659
(148), Pl. 17.
Stamp: Emperor’s bust and inscriptions: (…)OΛ(…).16
Occurence: Alexandria, Kellia (Egypt), Istanbul, Yenikapı
(Turkey), Athens, Kos (Greece), Tocra (Libia), Geronissos
(Cyprus).
See González Cesteros in this volume.
108. Hammamet 3A, base, no. 1315 (135), Pl. 17.
Stamp: Palm branch, similar stamps in Hammamet golf.17
See Bonifay – Capelli in this volume.
123. Africana III A/B, neck, handle, no. 865 (190), Pl. 17.
Graffito cut after firing: K( ).
172. Miscellaneous, neck, no. 1356 (195), Pl. 15.
Stamp: [---]VR.
Date: Augustean-Julio-Claudian period.
Comment: Marseille, it is a Gauloise 2 type.18
182. Miscellaneous, body fragment, no. 846 (199), Pl. 15.
Red dipinto: K( ).
183. Miscellaneous, body fragment, no. 734 (200), Pl. 15.
Graffito on the body cut after firing: Υε(…).
Ephesus amphora fragment, the type is uncertain (one-han-
dled jar, LR 3 or Ephesus 56?).
184. Miscellaneous, body fragment, no. 38 (26) – 2130,
Pl. 15.
Nominal graffito on the body cut after firing:
IVLIVSSIFA(…).
The problem is how to cut the words: Iuli Ussi Fa[---] ? or
Iulius Sifa[---] ?
Is it a Roman graffiti?
16
Diamanti 2010, 8.
17
Bonifay 2004, 20, Fig. 8, 17-19.
18
Similar stamps in Laubenheimer 1985 and Bertucci 1991. There
is a VAL stamp in Nijmegen.
18.03.19 14:39
 79

9 C ATALOGUE
Tamás Bezeczky – Piero Berni Millet – Michel Bonifay –
Claudio Capelli – Horacio González Cesteros – Sándor Józsa –
György Szakmány

ABBREVIATIONS
Name - simple name DNB - diameter at the junction of neck and body
No. - number or inventory number DB - diameter of the body
Stamp - dominus stamp DF - diameter of the foot
Stamp* - vilicus/workshop manager stamp H - height of the fragment
Dim - dimension measured HT - height of the amphora
D - diameter of the rim S - section of the handle
V - thickness of the rim Colour - Munsell Colour
HD - height of the rim SF - surface colour
DN - minimum diameter of the neck VC - visual characteristics
HN - height of the neck Date - date range

ITALIAN AND ADRIATIC

Lamboglia 2 Dim: D = 18 cm, V = 3 cm, HD = 3.4 cm, H = 5.5 cm
Colour: reddish yellow (5YR 6/6)
1. VC: Hard, fine-textured fabric, sparse scatter of small red iron ore, few
Name: Rim, neck reaction rims surrounding voids which held limestone.
No.: 18 (09) - 3561 Date: first quarter of the first century B.C.
Context: Room Cist.X Origin: Italy, W coast of the Adriatic
Dim: D = 17 cm, V = 2.9 cm, HD = 4.3 cm, H = 10.4 cm
Colour: light red (2.5YR 6/6) Dressel 6A
VC: Hard, very fine-grained, micaceous fabric. Sparse scatter of small red
iron ore and fine-grained limestone particles. 4.
Date: late second to late first century B.C. Name: Rim
Origin: Italy, W coast of the Adriatic No.: 18 (10) - 3563
Context: Room?, 128
2. Dim: D = 19 cm, V = 3.2 cm, HD = 5.4 cm, H = 8.3 cm
Name: Rim, neck Colour: buff (5YR 7/3)
No.: 533 (152) - 3960 VC: Hard, fine-textured fabric.
Context: Room H (RR) Date: late first century B.C. to mid-first century A.D.
Stamp: BAR·ME(…), on the rim. This stamp has not been found else- Origin: Italy, W coast of the Adriatic
where.
Dim: D = 16 cm, V = 2.5 cm, HD = 5.1 cm, H = 9.2 cm 5.
Colour: light reddish brown (5YR 6/6), SF: buff (7.5YR 7/4) Name: Rim, neck
VC: Hard, fine-textured fabric, slightly micaceous, sparse scatter of small No.: 140 (54) - 25973
red iron ore. Context: Courtyard, 106
Date: late second to late first century B.C. Stamp: V, V(…), on the rim
Origin: Italy, W coast of the Adriatic Dim: D = 18 cm, V = 2.4 cm, HD = 4 cm, H = 7.5 cm
Colour: reddish yellow (5YR 6/6)
3. VC: Hard, sandy fabric. Scatter of small red iron ore and fine-grained
Name: Rim limestone particles.
No.: 656 (118) Date: late first century B.C. to mid-first century A.D.
Context: ? Origin: Italy, W coast of the Adriatic
Stamp: VERSO, on the rim

BEZECZKY_Text.indd 79 18.03.19 14:39

 80 THE AMPHORAE

6. Dressel 6B
Name: Base
No.: 18 (11) - 3567 Laecanius production, Fažana
Context: Room?, 128
Dim: DF = 3.9 cm, H = 25 cm 11.
Colour: buff (5YR 7/4) Name: Rim frgm.
VC: Hard, fine-textured fabric. Sparse scatter of small red iron ore, few No.: 1493 (01)
reaction rims surrounding voids which held limestone, microfossil. Context: Room 203 C
Date: late first century B.C. to mid-first century A.D. Stamp: ..EK·A, (C) [La]ek(ani) (Bassi), A(...).
Origin: Italy, W coast of the Adriatic Dim: D = 15 cm, V = 2.1 cm, H = 13.7 cm
Colour: light red - reddish yellow (2.5 YR 6/6 - 5 YR 6/8)
7. VC: Hard, very fine-grained fabric. Frequently scattered rounded carbon-
Name: Rim, neck ate particles, and microfossils, shells.
No.: 140 (153) - 25972 Thin-section: Massive brownish red matrix. Non-carbonatics: very well
Context: Courtyard, 106 sorted, very fine-grained (15-70 μm), medium amount, mainly
Graffito: XXIIII on the rim, post cocturam quartz. Micritic carbonate: abundant, coarser-grained (40-200 μm),
Dim: D = 16.5 cm, V = 2.4 cm, HD = 4.7 cm, H = 6.5 cm well sorted, rounded. Many globigerinoids. Few micas, limonitic con-
Colour: light red (2.5YR 6/6), SF: buff (5YR 7/4) cretions, bivalves, forams, opalic spiculae, terra rossa. Raw materials:
VC: Hard, fine-textured fabric. Sparse scatter of small red iron ore and mainly flysch, little terra rossa, little recent marine material.
fine-grained limestone particles, microfossil. Date: Tiberian / early Claudian period
Date: late first century B.C. to mid-first century A.D.
Origin: Italy, W coast of the Adriatic 12.
Name: Rim-, neck frgm., handle
8. No.: 38 (25) - 2129
Name: Rim, neck, handle Context: Room D
No.: 1025 (106) Stamp*: .YNT, [S]ynt(...)
Context: ? Dim: D = 14 cm, H = 10.4 cm.
Dim: D = 16.5 cm, V = 2.2 cm, HD = 4.3 cm, DN = 11 cm, S = 3.8 × Colour: light red (2.5 YR 6/6)
4.4 cm, H = 39 cm VC: Hard, very fine-grained fabric. Frequently scattered rounded lime-
Colour: reddish yellow (5YR 6/6) stone particles, and frequent microfossils.
VC: Hard, fine-textured fabric, voids, sparse scatter of small red iron ore Thin-section: Massive brownish red matrix, few elongated holes. Non-car-
and fine-grained limestone particles. bonatics: very well sorted very fine-grained (15-80 μm), large amount,
Date: late first century B.C. to mid-first century A.D. mainly quartz. Micritic carbonate: abundant, coarser-grained (40-
Origin: Italy, W coast of the Adriatic 200 μm), well sorted, rounded. Many globigerinoids, accessory min-
erals. Few micas, bivalves, forams, limonitic concretions, opalic spicu-
Dressel 6B earliest form (1A phase) / ante Dressel 6B lae, terra rossa. Raw materials: mainly flysch, little terra rossa, little
recent marine material.
9. Date: Tiberian / early Claudian period
Name: Rim, neck, handle
No.: 221 (61) - 3742 13.
Context: Cist. X, 192 Name: Rim frgm.
Dim: D = 19.5 cm, V = 3.2 cm, HD = 4.6 cm, H = 17.2 cm No.: 1326 (611)
Colour: reddish yellow (5YR 6/6) Context: 221
VC: Hard, very fine-grained, slightly micaceous fabric. Sparse scatter of Stamp*: COMVS, Comus, S retro
small, fine grained limestone particles. Dim: D = 15 cm, V = 2.2 cm, H = 6.9 cm
Thin-section: Massive brownish red matrix, different sized holes. Non-car- Colour: reddish yellow - yellowish red (5 YR 6/6 - 5/6)
bonatics: moderately sorted, very fine-grained (15-60 μm), medium VC: Hard, very fine-grained fabric. Frequently scattered rounded lime-
amount, few coarser (100-180 μm) mainly quartz. Many micas, ac- stone particles, and frequent microfossils.
cessory minerals. Few globigerinoids (?), limonitic concretion, terra Thin-section: Massive brownish red matrix, few large holes. Non-carbo-
rossa. Raw materials: mainly clayey flysch, more terra rossa. natics: very well sorted, very fine-grained (15-80 μm), low amount,
Date: mid-first to the end of the first century B.C. mainly quartz. Micritic carbonate: abundant, coarser-grained (40-
Origin: North Italy or Istria? 200 μm), well sorted, rounded. Many micas, globigerinoids, limonitic
concretions. Few opalic spiculae, terra rossa. Raw materials: mainly
10. flysch, little terra rossa, little recent marine material.
Name: Base Date: Tiberian / early Claudian period
No.: 69 (42) - 3898
Context: Room K, 190 14.
Dim: FD = 8.1 cm, H = 8 cm Name: Rim
Colour: yellowish red (5YR 5/6) No.: 1243 (110)
VC: Hard, very fine-grained fabric. Frequently scattered rounded lime- Context: 101
stone inclusions, and microfossils. Stamp: LAE, (C) LAE(kani) (Bassi)
Date: mid-first to the end of the first century B.C. Dim: D = 14.5 cm, V = 1.8 cm, H = 4.8 cm
Origin: North Italy or Istria? Colour: reddish yellow (5YR 7/6), SF: light red (2.5YR 6/6)

BEZECZKY_Text.indd 80 18.03.19 14:39

 CHAPTER 9 – Catalogue (Bezeczky et al.) 81

VC: Very fine-grained fabric, a dissolved secondary carbonate. Imperial production, Fažana
Thin-section: Mixture of non-calcareous brownish red and calcareous
brownish grey matrix. Non-carbonatics: abundant, well sorted, very 18.
fine-grained (15-100 μm), mainly quartz. Micritic carbonate: few, Name: Rim frgm.
coarse-grained (generally 100-400 μm maximum 1000 μm). Many No.: 1476 (648)
globigerinoids. Few micas, limonitic concretions, terra rossa. Raw Context: 203 B
materials: limy and non-calcareous clayey flysch, very little terra rossa, Stamp: IM[P], Im[p](eratoris)
very little recent marine material. Dim: D = 14 cm, V = 2.3 cm, H = 7.5 cm
Date: Claudian / Flavian period Colour: reddish yellow (5 YR 6/6)
VC: Soft, fine-grained fabric, scattered rounded limestone particles, and
15. microfossils.
Name: Rim Thin-section: Massive brownish red matrix, few large holes. Non-carbonat-
No.: 719 (117) ics: very well sorted, very fine-grained (15-70 μm), medium amount,
Context: ? few coarser-grained (100-140 μm) mainly quartz. Micritic carbonate:
Stamp: C.L…, C. L[aek](ani) (Bassi) abundant, coarser-grained (40-200 μm), well sorted, rounded. More
Dim: D = 14 cm, V = 1.9 cm, H = 8.1 cm limonitic concretions. Few micas, globigerinoids, recent foram, opalic
Colour: reddish yellow (5YR 6/6) spiculae. Raw materials: mainly flysch, very little terra rossa, little re-
VC: Soft, larger iron ore particles, and microfossils, few reaction rims sur- cent marine material.
rounding voids which held limestone, irregularly fired. Date: A.D. 78/81 - 81
Thin-section: Massive brownish red matrix. Non-carbonatics: very well
sorted, very fine-grained (15-60 μm), medium amount. Micritic 19.
carbonate: coarser-grained, medium amount. More micas, globigeri- Name: Rim
noids, large bivalves, limonitic concretion, terra rossa. Raw materials: No.: 1338 (120)
mainly flysch, more terra rossa, very little recent marine material. Context: 111-112
Date: Claudian / Flavian period Stamp: IMP·TRA, Imp(eratoris) Tra(iani), upside down
Stamp*: ...VS.
16. Dim: D = 14 cm, V = 1.9 cm, HD = 5 cm, H = 6.2 cm
Name: Rim Colour: light red (2.5YR 6/8)
No.: 1279 (123) VC: Hard fabric, scattered rounded limestone particles, iron ore, angular
Context: 103B voids.
Stamp: …EK, (C) [LA]EK(ani) (Bassi) Thin-section: Massive inhomogeneous brownish red and darker red-
Dim: D = 14 cm, V = 1.9 cm, H = 6.8 cm dish brown matrix, few various-sized holes. Non-carbonatics: very
Colour: reddish yellow (5YR 6/6) well sorted, very fine-grained (15-60 μm), medium amount, mainly
VC: Soft fabric, iron ore particles, few reaction rims surrounding voids quartz. Micritic carbonate: abundant, coarser-grained (40-150 μm),
which held limestone. well sorted, few coarse (generally 250-400 μm) rounded. More micas,
Thin-section: Massive brownish red matrix. Non-carbonatics: very well globigerinoids, opalic spiculae. Few forams, limonitic concretions.
sorted, very fine-grained (15-60 μm), medium amount, few coarser Raw materials: mainly flysch, very little terra rossa, little recent ma-
(100-140 μm) mainly quartz. Micritic carbonate: coarser-grained rine material.
(40-200 μm), well sorted, rounded. Many globigerinoids. Few micas, Date: Trajanic period
limonitic concretions, opalic spiculae. Raw materials: mainly flysch,
little terra rossa, little recent marine material. 20.
Date: Claudian / Flavian period Name: Rim
No.: 1426 (122)
17. Context: 209?
Name: Rim Stamp: ..P.TRA, [Im]p(eratoris) Tra(iani), upside down
No.: 1371 (121) Dim: D = 15 cm, V = 2 cm, H = 6.1 cm
Context: ? Colour: reddish yellow (5YR 6/8)
Stamp*: PA…, Pa[gani] VC: Soft fabric, scattered fine brown, white particles.
Dim: D = 14 cm, V = 2 cm, H = 6 cm Thin-section: Massive brownish red matrix. Non-carbonatics: very well
Colour: reddish yellow (5YR 6/8) sorted, very fine-grained (15-70 μm), large amount, few coarser
VC: Soft fabric, scattered rounded limestone and iron ore particles. (100-140 μm) mainly quartz. Micritic carbonate: coarser (generally
Thin-section: Non-massive, inhomogeneous pale brownish red and darker 40-200 μm), abundant, well sorted, rounded. More globigerinoids,
reddish brown matrix, much fine porous. Non-carbonatics: very well opalic spiculae. Few micas, forams, molluscs, limonitic concretions.
sorted, very fine-grained (15-60 μm), large amount, few coarser (100- Raw materials: mainly flysch with very little terra rossa, little recent
140 μm), mainly quartz. Few globigerinoids, limonitic concretion, marine material.
terra rossa. More micas, opalic spiculae. Raw materials: mainly flysch Date: Trajanic period
mixed with some terra rossa, little recent marine material.
Date: Claudian / Flavian period 21.
Name: Rim
No.: 12 (116) - 26017
Context: 143
Stamp: .MP.T.., [I]mp(eratoris) T[ra](iani)

BEZECZKY_Text.indd 81 18.03.19 14:39

 82 THE AMPHORAE

Dim: D = 14 cm, V = 1.9 cm, H = 5 cm Fažana 1, Fažana, Istria

Colour: light red (2.5YR 6/8)
VC: Soft fabric, scattered bigger black and brown, white particles. 26.
Thin-section: Massive brownish red matrix. Non-carbonatics: very well Name: Rim, neck, handle
sorted, very fine-grained (15-70 μm), medium amount, mainly No.: 1 (1) - 16286
quartz. Micritic carbonate: coarser-grained (40-150 μm), abundant, Context: 104 ?
few large (100-400 μm), well sorted, rounded. More micas, globi- Dim: D = 10.5 cm, V = 1.6 cm, HD = 7.6 cm, H = 14.8 cm
gerinoids. Few limonitic concretions, terra rossa, opalic spiculae. Raw Colour: reddish yellow (5YR 7/6)
materials: mainly flysch, little terra rossa, little recent marine material. VC: Hard fabric, scattered fine brown, white particles, few reaction rims
Date: Trajanic period surrounding voids which held limestone.
Date: last third of the second to early third century A.D.
22.
Name: Rim Fažana 2, Fažana, Istria
No.: 500 (119) - A 10627
Context: 447 27.
Stamp*: LESBI (reverse). Name: Rim
Dim: D = 16.5 cm, V = 2.5 cm, H = 6.7 cm No.: 1370 (164)
Colour: light red (2.5YR 6/8) Context: Dolia 4
VC: Very hard, micaceous, very fine fabric, scattered fine brown, white Dim: D = 10.3 cm, V = 1.4 cm, H = 10.4 cm
particles. Colour: reddish yellow (5YR 6/8)
Thin-section: Massive inhomogeneous brownish red and darker red- VC: Hard fabric, scattered bigger dark brown and white particles and iron
dish brown matrix, few elongated holes. Non-carbonatics: very well ore.
sorted, very fine-grained (15-80 μm), large amount, few coarser Date: last third of the second to end of the third century A.D.
(100-300 μm), mainly quartz. Micritic carbonate: coarser-grained
(40-100 μm), abundant, well sorted, rounded. Few micas, globigeri- Porto Recanati
noids, molluscs, limonitic concretions, terra rossa, opalic spiculae.
Raw materials: mainly flysch with little terra rossa, little recent ma- 28.
rine material. Name: Rim
Date: Domitianic period No.: 994 (104)
Context: ?
23. Dim: D = 16 cm, V = 2.5 cm, H = 13.5 cm
Name: Rim, neck Colour: red (2.5YR 5/8)
No.: 1347 (124) VC: Hard, fine, frequent microfossils, scattered brown and white particles.
Context: Room behind the press Date: from the early Claudian period to the second century A.D.
Dim: D = 15 cm, V = 2 cm, HD = 5 cm, H = 10.5 cm Origin: Italy, W coast of the Adriatic
Colour: reddish yellow (5YR 7/6)
VC: Soft, fine fabric, scattered fine brown, microfossils, few reaction rims 29.
surrounding voids which held limestone, microfossil. Name: Rim, neck, handle
Date: Tiberian to Hadrianic period No.: 1419 (151)
Context: ?
24. Dim: D = 13 cm, V = 1.5 cm, HD = 8.3 cm, H = 28 cm
Name: Rim, base Colour: light reddish yellow (5YR 6/6)
No.: 85 (46) - 3390 and 3381 VC: Hard, fine fabric, scattered bigger brown and white particles, differ-
Context: Room K ent fabric from the other Porto Recanati amphora no. 28.
Dim: D = 15 cm, V = 2.2 cm, HD = 5 cm, DF = 4.5 cm Date: from the early Claudian period to the second century A.D.
Colour: red (2.5YR 5/6) Origin: Italy, W coast of the Adriatic
VC: Hard, fine, Strange Fažana fabric shows frequent white and brown-
ish particles, microfossils, few reaction rims surrounding voids which Forlimpopoli
held limestone, microfossil.
Date: Tiberian to Hadrianic period 30.
Name: Neck, handle
25. No.: 623 (80)
Name: Rim, neck, handle Context: ?
No.: 104 (49) - 23032 Dim: D = 9.3 cm, V = 1.7 cm, HD = 1.2 cm, H = 12.7 cm
Context: 37? Colour: very pale brown (10YR 7/4)
Dim: D = 14.6 cm, V = 2.1 cm, HD = 5 cm, S = 3 × 3.1 cm, H = 27.5 cm VC: Hard, fine fabric.
Colour: red (2.5YR 5/8) Thin-section: Dark brownish grey, partly fired calcareous and matrix. Non-
VC: Soft fabric shows frequent white and brownish particles, frequent carbonatics: moderately sorted, abundant, very fine - fine-grained
microfossils, shells. (generally 15-100 μm), few coarser (until 300 μm) mainly quartz.
Date: Tiberian to Hadrianic period Much micas, strongly fired biotite. Carbonate: more fine-grained
strongly fired, few large (250-700 μm) rounded silty limestone clasts
(flysch).

BEZECZKY_Text.indd 82 18.03.19 14:39

 CHAPTER 9 – Catalogue (Bezeczky et al.) 83

Date: first century to the end of the third century A.D. Dressel 2-4
Origin: Italy, W coast of the Adriatic
34.
31. Name: Handle
Name: Neck, handle No.: 31 (15) - 1489
No.: 1267 (170) Context: Street, 137
Context: Room 102 Dim: S = 2.3 × 4.7 cm, H = 16.2 cm
Dim: D = 9 cm, V = 1.1 cm, HD = 1.2 cm, S = 1.9 × 5.4 cm, H = 8.1 cm Colour: buff (7.5YR 7/4)
Colour: buff (10YR 8/4) VC: Hard, fine fabric, iron ore and limestone inclusions, frequent white
VC: Hard, fine fabric, small reddish inclusion, mica, small voids. reaction rims around voids, indicating where the limestone grains
Date: first century to the end of the third century A.D. once existed.
Origin: Italy, W coast of the Adriatic Origin: Italy, Adriatic region

Aquincum 78 35.
Name: Handle
32. No.: 386 (87) - 25128
Name: Rim Context: Room GL
No.: 1527 (163) Dim: S = 2 × 4.5 cm, H = 16.3 cm
Context: Room 219 Colour: red (2.5YR 5/6)
Dim: D = 10 cm, V = 1 cm, HD = 2.6 cm, H = 7 cm VC: Campanian, black sand fabric.
Colour: buff (10YR 8/4) Date: first century A.D.
VC: Hard, very fine inclusions. Origin: Campania, Vesuv region
Date: mid-first century to the end of the third century A.D.
Origin: Istria? 36.
Name: Rim, neck, handle
33. No.: 1308 (142)
Name: Rim Context: Room 105/B2
No.: 1310 (127) Dim: D = 12 cm, V = 1 cm, S = 1.8 × 3.2 cm, H = 14.5 cm
Context: Room 221 Colour: reddish yellow (5YR 6/6), SF: reddish yellow (5YR 7/6)
Dim: D = 8.8 cm, V = 1.1 cm, H = 4.1 cm VC: Hard, very fine fabric. Red iron ore, brownish inclusions, irregularly
Colour: buff (10YR 7/3) fired.
VC: Hard, very fine, some bigger iron ore and limestone inclusions. Date: first to third century A.D.
Date: mid-first century to the end of the third century A.D. Origin: Italy, Adriatic region
Origin: Istria?

SPANISH
Dressel 7-11 Dressel 14A

37. 39.
Name: Rim Name: Rim, neck, handle
No.: 455 (66) - 9251 No.: 1341 (157)
Context: Room A, 111, Context: Room 231 Q
Dim: D = 15.5 cm, V = 3.5 cm, HD = 2.8 cm, H = 4.2 cm Dim: D = 19.8 cm, V = 2.2 cm, S = 4 × 6.6 cm, H = 6.1 cm
Colour: buff (7.5YR7/4), SF: white (2.5Y 8/2) Colour: buff (7.5YR 7/4)
VC: Cádiz fabric, sandy, hard. VC: Hard and slightly coarse fabric.
Date: second half of the first century A.D. Date: late first and second century A.D.
Origin: Cádiz region Origin: Málaga?

Beltran IIA Dressel 20

38. 40.
Name: Base Name: Handle
No.: 10 (6) - 16027 No.: 1271 (125)
Context: Room?, 156 Context: Room 215
Graffito: V (…), post cocturam Stamp: palm branch M(…), ((ramus palmae)) M[---].
Dim: DF = 3.8 cm, V = 0.8 cm, H = 9.1 cm Dim: S = 4.3 cm, H = 12.3 cm
Colour: buff (5YR 7/2) Colour: light reddish brown (5YR 6/4)
VC: Hard, fine fabric, fine red, colourless and white inclusions, mica. VC: Guadalquivir, very hard, coarse.
Date: second half of the first century A.D. and second century A.D. Date: first half of the second century A.D. The palm branch was in vogue
Origin: Cádiz region during Hadrian’s period. Martin-Kilcher 1987, 94, ST 2a, MAF ((ra-
mus palmae)) on Antoninian handle (Hadrian or Antoninus Pius).
Origin: Guadalquivir Valley

BEZECZKY_Text.indd 83 18.03.19 14:39

 84 THE AMPHORAE

41. Almagro 51A-B

Name: Body, base
No.: 1554 (115) 47.
Context: ? Name: Rim
Graffito: Asiatici / sec( ) x k(alendas) mar(tias) Alb( ) + II, ante cocturam No.: 140 (52) - 25940
Dim: DB = 52 cm, V = 2.5 cm, H = 58.5 cm Context: Courtyard, 106
Colour: light brown to light grey (7.5YR 6/4 to 6/2) Dim: D = 8.9 cm, V = 1.3 cm, HD = 2.8 cm, H = 4 cm
VC: Hard, medium-grained Guadalquivir fabric. Colour: red (2.5YR 5/6), SF: buff (7.5YR 7/4)
Date: second third of the second century A.D. VC: Lusitania (Tejo/Sado or Algarbe) hard, coarse fabric, reddish, dark
Origin: Guadalquivir Valley brown and colourless inclusions. The microscope shows frequent
white reaction rims around voids, indicating where limestone grains
42. once existed. Coated.
Name: Rim Date: late fourth and fifth century A.D.
No.: 1378 (156) Origin: Lusitania
Context: Room 204
Dim: D = 20 cm, V = 4.4 cm, H = 4.5 cm 48.
Colour: light brown (7.5YR 6/4) Name: Rim
VC: Guadalquivir, very hard, coarse fabric. No.: 140 (168) - 25939
Date: second third of the second century A.D. Context: Courtyard, 106
Origin: Guadalquivir Valley Dim: D = 8 cm, V = 0.9 cm, HD = 0.8 cm, H = 7.1 cm
Colour: buff (7.5YR 7/4)
43. VC: Lusitania (Tejo/Sado or Algarbe) soft, fine, sandy fabric, fine reddish
Name: Rim, neck inclusion, some voids.
No.: 923 (89) Date: fifth century A.D.
Context: Room 220 Origin: Lusitania
Dim: D = 16 cm, V = 3.8 cm, HD = 3.2 cm, H = 5 cm
Colour: light reddish brown (5YR 6/3) Almagro 51C
VC: Guadalquivir, very hard, coarse fabric.
Date: end of the second-beginning of the third century A.D. 49.
Origin: Guadalquivir Valley Name: Rim, neck, handle
No.: 34 (21) - 16159
44. Context: 36?
Name: Rim, neck Dim: D = 11 cm, V = 1.5 cm, H = 4.3 cm
No.: 1391 (126+201) other part 1375 box Colour: dark buff (7.5YR 7/2)
Context: Room 233, Garden VC: Lusitanian hard, coarse, sandy fabric, red and colourless inclusions.
Dim: D = 19 cm, V = 3 cm, H = 8.2 cm Irregularly fired.
Colour: reddish yellow (5 YR 7/6). Date: fourth century A.D.
VC: Guadalquivir, hard, coarse fabric. Big quartz inclusions. Origin: Lusitania
Date: mid-third century A.D.
Origin: Guadalquivir Valley Beltran 68

45. 50.
Name: Rim Name: Rim, neck, handles
No.: 1407 (158) No.: 1221 (60)
Context: Room 234 Context: Courtyard
Dim: D = 15 cm, V = 2.7 cm, H = 5 cm Dim: D = 12 cm, V = 1.6 cm, HD = 3.2 cm, S = 2.9 × 5.2 cm, H = 15 cm
Colour: reddish yellow (7.5YR 6/6) Colour: red (2.5YR 5/4)
VC: Hard, fine, sandy fabric. VC: Hard, very fine-grained sandy fabric common with quartz inclusions,
Date: third century A.D. limestone or voids that once contained this material.
Origin: Guadalquivir Valley Thin-section: Ca-rich matrix. Very abundant, moderately sorted inclusions
(<0.5 mm), mainly composed of subangular to frequently rounded
Almagro 50A/ Keay XXII quartz, Ca-microfossils and subordinate limestone fragments.
Date: third to fourth century A.D.
46. Origin: Cádiz region
Name: Base
No.: 626 (81) 51.
Context: ? Name: Base, lower part of the body
Dim: DF = 4.9 cm, H = 8.8 cm No.: 110 (63) - 15988
Colour: buff (5YR 7/4) Context: ?
VC: Lusitanian very fine fabric, reddish, dark brown and colourless inclu- Dim: DF = 5.5 cm, H = 26 cm
sions. Colour: red (10R 5/8), SF: buff (10YR 8/3)
Date: fourth and fifth century A.D. VC: Cádiz hard, coarse fabric, white and colourless inclusions.
Origin: Lusitania Date: third to late fourth/fifth century A.D.
Origin: Cádiz region
Comment: Probably same piece as no. 50.

BEZECZKY_Text.indd 84 18.03.19 14:39

 CHAPTER 9 – Catalogue (Bezeczky et al.) 85

GAULISH

Gauloise 4

52.
Name: Handle
No.: 632 (85)
Context: ?
Dim: S = 2 × 4.5 cm, H = 9.4 cm
Colour: reddish yellow (5YR 6/6)
VC: Very fine, hard, Narbonensis calcareous fabric, white and colourless
inclusions and angular voids.
Date: second half of the first to the mid-second century A.D.
Origin: Gallia Narbonensis

SICILIAN

Mid Roman 1

53. 54.
Name: Rim Name: Rim, neck, handle
No.: 207 (74) - 2339 No.: 1213 (169)
Context: Room J Context: Room AX-8/c
Dim: D = 9.5 cm, V = 1.2 cm, HD = 2 cm, H = 4.8 cm Dim: D = 7.4 cm, V = 1 cm, S = 2.1 × 2.8 cm, H = 6.7 cm
Colour: red (10R 5/6), SF: light brown (7.5YR 6/4) Colour: red (2.5YR 5/6)
VC: Hard, fine, calcareous fabric, white, black and colourless inclusions, VC: Similar to no. 53, but irregularly fired.
frequent limestone or white reaction rims surrounding voids that Date: ?
once contained this material. Origin: ?
Origin: ?

ITALIAN OR EASTERN MEDITERRANEAN?

Palatine East 1 56.

Name: Rim, neck, handle
55. No.: 453 (191) - 10395
Name: Rim, complete neck and beginning both handles Context: Room A
No.: 59 (196) Dim: D = 8.4 cm, V = 1 cm, HD = 1.2 cm, H = 11.5 cm
Context: Room D Colour: red (10R 5/6)
Dim: D = 9.2 cm, V = 1.5 cm, S = 2.5 × 3.7 cm, H = 13.6 cm VC: Very hard and very fine, calcareous fabric. Frequent microfossils, the
Colour: light red (10R 6/8), SF: buff (7.5YR 7/4) microscope shows frequent white reaction rims around voids, indicat-
VC: Very hard and fine calcareous fabric. Sparse, coarse, colourless and ing where limestone grains once existed. Some angular small quartz
light red inclusions and angular voids. Greenish slip (coated). and colourless inclusions.
Date: 5-6 A.D. Date: ?
Origin: ? Origin: ?

EASTERN MEDITERRANEAN

Crete TRC 2 (Portale – Romeo 2000) Dim: D = 7 cm, V = 1.1 cm, HD = 2.1 cm, S = 1.9 × 2.9 cm, H = 11.3 cm
Colour: light reddish brown (5YR 6/4)
57. VC: Very hard, not really coarse and fine. There are white limestone, red,
Name: Rim, neck, handle black and colourless inclusions and some angular voids.
No.: 31 (16) - 1491 Date: fifth to seventh century A.D.
Context: Street Origin: Crete

BEZECZKY_Text.indd 85 18.03.19 14:39

 86 THE AMPHORAE

Rhodian Koan

58. 63.
Name: Handle Name: Rim, neck, handle
No.: 982 (90) No.: 33 (19) - 1581
Context: ? Context: Room S, on the level of the doorstep
Dim: S = 2 cm, H = 19.3 cm Dim: D = 12 cm, V = 1.4 cm, HD = 1.6 cm, H = 8.6 cm
Colour: red (2.5YR 5/6) Colour: red (10YR 5/6)
VC: Hard, coarse fabric. Dark grey, colourless inclusions, some microfos- VC: Hard, fine fabric, moderate grey, dark red and colourless inclusions.
sil, the microscope shows frequent white reaction rims around voids, The microscope shows frequent white reaction rims around voids, in-
indicating where limestone grains once existed. dicating where limestone grains once existed.
Date: end of the first and second century A.D. Date: first to second century A.D.
Origin: Rhodes or the Rhodian Peraea, and near Caria on the mainland Origin: Kos

59. 64.
Name: Handle Name: Handle
No.: 926 (93) No.: 1230 (101)
Context: Room 209 Context: Room LJ, close to the sarcophagus
Dim: S = 2.6 × 2.9 cm, H = 9.7 cm Dim: S = 2.3 × 4.2 cm, H = 16.6 cm
Colour: reddish yellow (7.5YR 6/6) Colour: light red (2.5YR 6/6)
VC: Hard, very fine calcareous fabric. Grey, red and colourless inclusions. VC: Hard, fine fabric, sparse peppered with white angular inclusions. Few
Date: end of the first and second century A.D. shells.
Origin: Rhodes or the Rhodian Peraea, and near Caria on the mainland Date: first to second century A.D.
Origin: Kos
60.
Name: Handle 65.
No.: 1423 (160) Name: Base
Context: Room 236 No.: 840 (95)
Dim: S = 2.8 × 3 cm, H = 6 cm Context: Room A, quadrant 3
Colour: red (2.5YR 4/8) Dim: DF = 2.2 cm, H = 8 cm
VC: Hard and coarse fabric. Frequent medium and small limestone, dark Colour: red or light red (10R 6/8-5/8), SF: white-cream coated
red and colourless inclusions and voids. VC: Very hard, fine fabric, sparse peppered with white angular inclusions.
Date: first and second century A.D. Few microfossils. The microscope shows frequent white reaction rims
Origin: Rhodes or the Rhodian Peraea, and near Caria on the mainland around voids, indicating where limestone grains once existed.
Date: Augustan Period?
Knidian Origin: Kos

61. Dressel 5
Name: Base
No.: 90 (47) - 2882 66.
Context: Room R, 13? Name: Handle
Dim: DF = 5 cm, H = 10 cm No.: 629 (78)
Colour: red (2.5YR 5/6) Context: Room 213
VC: Hard and fine fabric. Moderate red and colourless, rarely limestone Dim: S = 2.4 × 4.5 cm, H = 6 cm
inclusions. Colour: white (10YR 8/2)
Date: first century A.D. VC: Hard, coarse fabric. Moderate coarse black and colourless inclusions.
Origin: Knidos Macro- or meso-sized voids. Probably lime/calcareous fabric, fired at
a high temperature.
Koan type with pinched handle Date: first century A.D.
Origin: South Aegean
62.
Name: Neck, handle Crétoise AC 4
No.: 1327 (155)
Context: Room 221 67.
Dim: S = 2.3 × 4.5 cm, H = 7.5 cm Name: Handle
Colour: red (2.5YR 5/6) No.: 1328 (161)
VC: Hard and fine, calcareous fabric, grey and red inclusions, frequent Context: Room 203 C
limestone or white reaction rims surrounding voids that once con- Dim: S = 2.6 × 3.2 cm, H = 6.8 cm
tained this material. White coated. Colour: buff (7.5YR 7/4)
Date: first century A.D. VC: Hard, fine sandy fabric. Moderate white, colourless and rare red in-
Origin: Kos or other Greek Islands clusions. Rare microfossils.
Date: first to second century A.D.
Origin: Crete. González Cesteros considers it a Rhodian type and dated it
in the Augustan period.

BEZECZKY_Text.indd 86 18.03.19 14:39

 CHAPTER 9 – Catalogue (Bezeczky et al.) 87

68. Agora M 279 similis

Name: Handle
No.: 1435 (162) 73.
Context: Room DR Name: Rim
Dim: S = 2.5 × 2.8 cm, H = 6.5 cm No.: 118 (167) - 16182
Colour: buff (7.5YR 7/4) Context: ?
VC: Hard and fine sandy fabric. Moderate grey and colourless inclusions. Dim: D = 3.7 cm, V = 0.5 cm, HD = 1.2 cm, H = 8 cm
Some shells. Colour: red (2.5YR 5/8)
Date: from the first century to early second century A.D. VC: Very hard and smooth fabric. Macro- and meso-sized inclusions,
Origin: Crete some angular quartz and some black and occasional small red inclu-
sions.
69. Date: second half third century A.D.
Name: handle Origin: not Ephesian production
No: 868 (205)
Context: Room A Agora M 126
Dim: S = 2.5 × 1.7 cm, H = 13.2 cm
Colour: light red (2.5YR 6/6) 74.
VC: Hard and very fine sandy fabric. Moderate white and colourless inclu- Name: Base
sions. No.: 1361 (134)
Date: from the first to early second century A.D. Context: Room 204
Origin: Crete Dim: DF = 5 cm, V = 0.6 cm, H = 6.5 cm
Colour: light red (2.5YR 6/6)
Agora G 199 / Pinched handle VC: Hard and fine micaceous fabric. Similar to no. 73, but more massive
and coarser. Rare white, red and colourless inclusions.
70. Date: second and beginning of the third century A.D.
Name: Handle Origin: Asia Minor
No.: 938 (97) Comment: There are no traces of dark grey slip. Typologically different to
Context: ? the Ephesian production, but also to other pieces like (133).
Dim: S = 2.5 × 2.8 cm, H = 5.7 cm
Colour: red (2.5YR 4/8) 75.
VC: Hard and smooth, fine fabric with white and colourless inclusions, Name: Base
no mica. No.: 1530 (165)
Date: end of the first to the start of the third century A.D. Context: Room 203
Origin: Cyprus or Cilicia Dim: DF = 4.7 cm, V = 0.7 cm, H = 5 cm
Colour: reddish yellow (7.5YR 6/6)
Agora F 66 VC: Hard and fine, calcareous fabric, white-greenish coated.
Date: late second to early third century A.D.
71. Origin: not Ephesian production
Name: Handle
No.: 10 (5) - 16038 76.
Context: Room ?, 156 Name: Base, body frgm. and handle frgm. (not illustrated)
Dim: S = 1.5 × 4 cm, H = 7.5 cm No.: 1370 (133)
Colour: red (2.5YR 5/8) Context: Dolia 4
VC: Hard fabric, quartz, white and grey inclusions, no mica. Dim: DF = 4.2 cm, V = 0.3 cm, H = 6.5 cm
Date: first century A.D. Colour: brownish yellow (10YR 6/8)
Origin: not Ephesian production VC: Hard, smooth, micaceous and calcareous fabric. Rare white and red
inclusions. With dark grey slip on of the outer wall and in the upper
Agora M 125 part of the body (handle) also in the inner wall.
Date: late second to early third century A.D.
72. Comment: Normally this slip is associated with the Pergamenian produc-
Name: Body, base tion.
No.: 1553 (150) - 15106
Context: ? Agora M 240
Dim: V = 0.4 cm, BD = 20.7 cm, DF = 3.2 cm, H = 43.2 cm
Colour: red (10R 4/6) 77.
VC: Ephesian hard, highly micaceous fabric, occasional fine-grained. Name: Rim, neck, handle
Origin: Ephesus region No.: 840 (94)
Context: Room A
Dim: D = 4 cm, V = 0.6 cm, HD = 0.5 cm, H = 5.8 cm
Colour: reddish yellow (7.5YR 6/6)
VC: Hard, micaceous and calcareous fabric. Sparse white and colourless
inclusions. Perhaps Pergamum fabric.
Date: third to fourth century A.D.
Origin: Pergamon?

BEZECZKY_Text.indd 87 18.03.19 14:39

 88 THE AMPHORAE

Kapitän 1 Dim: D = 9.3 cm, V = 1.1 cm, S = 2.4 × 3.3 cm, H = 14.2 cm
Colour: reddish yellow (5YR 6/6)
78. VC: Hard and relatively smooth fabric. Grey, black and colourless inclu-
Name: Base sions.
No.: 1237 (159) Date: end of fifth to sixth century A.D.
Context: Room Alfa Origin: Cilicia
Dim: H = 16.5 cm Comment: LRA 1B1
Colour: red (2.5YR 5/8)
VC: Medium hard and coarse fabric, laminated. White coated. With 83.
some small mica and occasional iron ore, some small semi-rounded Name: Rim, neck, handle (rim deformed)
colourless, black and dark grey inclusions. The fabric looks like No.: 108 (51) - 26139
Aegean-Asia Minor. Context: Garden
Date: second to fourth century A.D. Dim: D = 8.6 cm, V = 1.2 cm, S = 2.8 × 3.7 cm, H = 9.2 cm
Origin: Aegean-Asia Minor region Colour: red (10R 5/6)
VC: Hard, sandy fabric rarely with inclusions of limestone. Sparce red and
Kapitän 2 dark grey inclusions. Rarely Microfossils and shells.
Date: sixth century A.D.
79. Origin: Cilicia
Name: Handle Comment: LRA 1B1
No.: 1389 (154)
Context: Room 233 84.
Dim: S = 3.1 × 5 cm, H = 10.8 cm Name: Rim, neck, handle
Colour: red (2.5YR 5/8) No.: 140 (55) - 25946
VC: Hard, fine and smooth fabric. Moderate white, colourless and grey Context: Courtyard, 106
inclusions. Fabric not typical of Kapitän 2. Dim: D = 8.2 cm, V = 1.1 cm, S = 2.8 × 3.5 cm, H = 11 cm
Date: third and fourth century A.D. Colour: strong brown (7.5YR 5/8)
Origin: North Aegean or Black Sea region VC: Hard, coarse Cilician fabric often with inclusions of limestone. Mod-
Comment: We have another handle fragment (Box 1473) with a different erate red, grey, colourless and black inclusions.
type of fabric. It is brick red and more coarse, bigger with many more Date: late sixth century A.D.
inclusions. Origin: Cilicia
Comment: LRA 1B
Agora M 273
85.
80. Name: Rim, neck, handle
Name: Rim, neck, handle No.: 497 (149) - 24699
No.: 453 (69) - 9204 Context: ?
Context: Room A, 196 Dim: D = 10 cm, V = 1.3 cm, H = 11 cm
Dim: D = 13 cm, V = 1.7 cm, HD = 2.2 cm, H = 17 cm Colour: reddish yellow (7.5YR 6/6)
Colour: buff (10YR 7/3) VC: Side hard, calcareous fabric. Common white, red, grey and black in-
VC: Hard, fine and sandy fabric. Moderate red, sparse colourless inclu- clusions. the microscope shows abundant white reaction rims around
sions. voids, indicating where limestone grains once existed.
Date: fifth century A.D. Date: fifth to early seventh century A.D.
Origin: Aegean Origin: Soli workshop, Cilicia

Late Roman Amphora 1 (LR 1) 86.

Name: Rim, neck, handles
81. No.: 1342 (136)
Name: Rim, neck, handle Context: Room 230 B
No.: 31 (18) - 1490 Graffito: A…, post cocturam
Context: Street Dim: D = 7 cm, V = 1 cm, S = 2 × 3.1 cm, H = 9.8 cm
Dim: D = 9.5 cm, V = 1 cm, S = 2. 7 × 3.4 cm, H = 6.1 cm Colour: red (2.5YR 5/8), SF: light red (2.5YR 6/6)
Colour: brown (2.5YR 5/2) VC: Hard, fine fabric. Sparce red, colourless and grey inclusions.
VC: Hard and very coarse fabric. Cilician fabric, abundant meso quartz Date: sixth and early seventh century A.D.
inclusions and some macro lime and quartz, also some voids. Irregu- Origin: Cilicia or South Aegean
larly fired. Comment: LRA 1B2
Date: fifth and sixth century A.D.
Origin: Cilicia Late Roman Amphora 2 (LR 2)
Comment: LRA 1B
87.
82. Name: Rim, neck, handles
Name: Rim, neck, handle No.: 7 (4) - 16014
No.: 108 (57) - 26138 Context: Room ?, 185
Context: Garden

BEZECZKY_Text.indd 88 18.03.19 14:39

 CHAPTER 9 – Catalogue (Bezeczky et al.) 89

Dim: D = 11 cm, V = 1.5 cm, HD = 4.7 cm, S = 2 × 3.4 cm, H = 17.5 cm Dim: DF = 2.7 cm, H = 6.5 cm
Colour: reddish yellow (5YR 6/6) Colour: red (2.5YR 4/6)
VC: Hard, coarse fabric. Often scattered white limestone inclusions. VC: Hard, very fine-grained, micaceous, laminated Ephesian fabric.
Voids are sparce. Date: fifth and sixth century A.D.
Date: fifth and sixth century A.D. Origin: Ephesus region
Origin: Argolid fabric Comment: Small version Ephesus 56 type.

88. Agora M 307

Name: Rim, beginning of the neck
No.: Box 399 (192) - 16496 93.
Context: Room E Name: Rim, neck, handle
Dim: D = 11.5 cm, V = 1.6 cm, HD = 5.6 cm, H = 7 cm No.: 571 (107) - 16634
Colour: pink or light red (10R 6/6 - 6/8) Context: Garden
VC: Hard, fine, smooth fabric, white, red and rare grey inclusions and Dim: D = 3.6 cm, V = 0.5 cm, HD = 0.5 cm, S = 1.3 × 2 cm, H = 8.7 cm
sparce voids. Colour: red (2.5YR 4/6)
Date: fifth and sixth century A.D. VC: Hard, very fine, micaceous Ephesus fabric. Often scattered very fine
Origin: ? white limestone inclusions.
Comment: Similar to some Chian amphorae from the Imperial period. Date: fifth century A.D.
Probably the production place is in Chios or around the island. Origin: Ephesus region

Late Roman Amphora 3 (LR 3) 94.

Name: Base
89. No.: 571 (108) - 16638
Name: Rim, neck, handle Context: Garden
No.: B 73 (180) Dim: DF = 3 cm, V = 0.3 cm, H = 2 cm
Context: ? Colour: red (2.5YR 5/8)
Dim: D = 3.6 cm, V = 0.4 cm, S = 1.2 × 2.9 cm, H = 9.2 cm VC: Hard, very fine, micaceous Ephesus fabric.
Colour: red (2.5YR 4/8) Date: fifth century A.D.
VC: Hard, micaceous, very fine Ephesian fabric. Origin: Ephesus region
Date: end of the fourth to end of the sixth century A.D.
Origin: Ephesus region 95.
Name: Base
90. No.: 571 (109) - 16639
Name: Base Context: Garden
No.: 350 (70) - 9964 Dim: DF = 3 cm, V = 0.3 cm, H = 2.7 cm
Context: Room B2 Colour: red (2.5YR 5/8)
Dim: DF = 3.5 cm, V = 0.5 cm, H = 9.5 cm VC: Hard, very fine, micaceous Ephesus fabric.
Colour: red (2.5YR 4/8), SF: light red (2.5YR 6/4) Date: fifth century A.D.
VC: Hard, micaceous, very fine Ephesian fabric, with rare limestone inclu- Origin: Ephesus region
sions.
Date: end of the fourth to end of the sixth century A.D. 96.
Origin: Ephesus region Name: Rim, neck, handle, body, base restored
No.: B 72 (181)
Ephesus 56 / Ephesus 8 Context: ?
Dim: D = 3.8 cm, V = 0.4 cm, HD = 0.7 cm, S = 0.9 × 1.8 cm, H = 35 cm
91. Colour: red (2.5YR 5/8)
Name: Rim, neck, handle, base VC: Hard, very fine, micaceous Ephesus fabric.
No.: 91 (48) - 25937 Date: fifth century A.D.
Context: 182? Origin: Ephesus region
Dim: H = 12 cm, V = 1.6 cm, DH = 1.1 cm, DN = 10.7 cm, S = 2.3 ×
3 cm, DF = 1.9 cm, H = 13.4 cm Late Roman Amphora 4 (LR 4) / Gaza
Colour: red-brown (10R 5/6 - 4/6)
VC: Hard, coarse, micaceous Ephesian fabric. Moderate quartz and red 97.
inclusions. Rare meso-sized limestone inclusions. Name: Rim
Date: fifth and sixth century A.D. No.: 1210 (105) - 10400
Origin: Ephesus region Context: Room A
Comment: Larger version Ephesus 56 type. Dim: D = 10.8 cm, V = 0.8 cm, H = 9.6 cm
Colour: reddish brown (2.5YR 5/4)
92. VC: Hard, very fine, sandy fabric, rarely big voids. Irregularly fired.
Name: Base Date: ?
No.: 763 (91) Origin: Gaza region
Context: ?

BEZECZKY_Text.indd 89 18.03.19 14:39

 90 THE AMPHORAE

98. Beirut?
Name: Base
No.: 1 (13) - 3155 102.
Context: Room Beta, 154 Name: Rim, neck
Dim: DF = 4.7 cm, V = 0.8 cm, H = 4.7 cm No.: 44 (33) - 25871
Colour: yellowish red (5YR 5/6) Context: 31 ?
VC: Hard, fine, brownish-red fabric. Abundant meso rounded quartz in- Dim: D = 12 cm, V = 1.3 cm, HD = 2.2 cm, H = 7.7 cm
clusions. Colour: yellowish red (5YR 5/6)
Date: sixth to seventh century A.D. VC: Very hard, coarse fabric. Large amount of small and medium rounded
Origin: Gaza red and grey inclusions.
Comment: Small base with a lot of grooves and completely flat bottom Date: fourth century A.D.
Gaza Amphora. Origin: Beirut region?

99. Cretan: AC 1C/D, MRC 3 (Portale – Romeo 2000)

Name: Handle
No.: 26 (14) 103.
Context: Room?, 346 Name: Rim, neck
Dim: S = 1.7 × 2.7 cm, H = 6.6 cm No.: 1261 (130)
Colour: light red (2.5YR 6/6) Context: Room 21 ?
VC: Hard, fine fabric. Abundant meso rounded quartz inclusions. The Dim: D = 7.8 cm, V = 1.2 cm, H = 9.6 cm
fabric is not like the typical brown Gaza amphorae; it is probably Colour: reddish yellow (5 YR 7/8), SF: reddish yellow (7.5 YR 8/6)
from the Northern part of Palestine. The surface has some similarities VC: Hard, coarse, slightly sandy fabric. Mega and macro limestone inclu-
with the fabric from the “Beirut”, but the inclusions are smaller and it sions, and some micro to macro voids. Sparce microfossils and shells.
does not have as many blue or dark quartz inclusions. Date: end of the second/beginning of the third century to beginning of
Date: B3 from Pieri? This would mean seventh century A.D. Also possible the fifth century A.D.
B1 (end of the fifth to beginning of the sixth century A.D.). Origin: ?

Late Roman Amphora 13 (LR 13) Athenian Agora M 235/6

100. 104.
Name: Neck Name: Rim, neck, handle
No.: 659 (148) No.: 455 (64) - 5247
Context: ? Context: Room A, 111
Stamp: Emperor’s bust and inscriptions: …OΛ Dim: D = 11.5 cm, V = 1.9 cm, HD = 4 cm, H = 11.8 cm
Dim: V = 1 cm, H = 10.2 cm Colour: light red (2.5YR 6/6)
Colour: reddish yellow (5YR 6/6) VC: Hard, very fine-grained sandy fabric with red, grey and rarely me-
VC: Koan hard, fine micaceous fabric. Brown, dark grey and quartz inclu- dium size quartz inclusions.
sions. Date: fourth and fifth century A.D.
Date: late sixth century A.D. Origin: Crete or Greece
Origin: Kos
Late Roman Amphora 7 (LR 7)
Samos Cistern Type
105.
101. Name: Base
Name: Rim, neck, handle No.: 173 (56=193) - 2784
No.: 63 (40) - 16585 Context: ?
Context: 8 ? Dim: FD = 4 cm, H = 7.8 cm
Graffito: X (…), post cocturam Colour: strong brown (7.5YR 4/6)
Dim: D = 10.5 cm, V = 1.1 cm, S = 1.3 × 2.9 cm, H = 12 cm VC: Soft, fine fabric. Abundant voids (large, medium, small), some small
Colour: red (10R 5/8), SF: brown (7.5YR 5/4) lime and sparse mica inclusions.
VC: Very hard, fine fabric with micro-sized red and colourless inclusions. Date: end of the fourth to seventh century A.D.
Often scattered white limestone inclusions. Origin: Egypt, Nile Delta
Date: sixth century A.D. Comment: LRA 7, probably end of fourth or fifth century A.D.
Origin: Asia Minor or Aegean Islands

BEZECZKY_Text.indd 90 18.03.19 14:39

 CHAPTER 9 – Catalogue (Bezeczky et al.) 91

BLACK SEA

Sinope C type

106. 107.
Name: Rim, neck Name: Rim, neck
No.: 1303 (143) No.: 1313 (131)
Context: Courtyard II Context: Room 221
Dim: D = 7.9 cm, V = 1.3 cm, H = 6 cm Dim: D = 10.6 cm, V = 1.1 cm, H = 7.3 cm
Colour: red (2.5 YR 4/8 - 5/8) Colour: red (2.5 YR 4/8 - 5/8).
VC: Very hard, fine fabric, with sparse black and limestone inclusions and VC: Hard, calcareous, fine fabric. Sparce coarse black inclusions, Fabric 5,
microfossils. Cream coated. Kassab Tezgör 2010.
Date: fourth century A.D.? Date: fourth and fifth century A.D.
Origin: Black Sea, Sinope? Origin: Sinope
Comment: fragment of type Sinope C II

AFRICAN
Hammamet 3A Africana I variant?

108. 111.
Name: Base Name: Rim, neck, handle
No.: 1315 (135) No.: 1223 (103)
Context: Room 202 Context: Room D
Stamp: Palm branch, ante cocturam Dim: D = 12.1 cm, V = 2.1 cm, HD = 4.8 cm, H = 11.6 cm
Dim: DF = 7.5 cm, H = 10.2 cm Colour: red (10R 5/8)
Colour: red (2.5YR 5/6) VC: Hard, medium-grained sandy fabric with rare black, reddish and
VC: Hard, medium-grained sandy fabric, with visible quartz and lime- moderate quartz inclusions, occasional grog visible.
stone inclusions. Thin-section: Fe-rich clay matrix. Abundant inclusions, composed of an-
Thin-section: Fe-rich clay matrix. Moderately sorted inclusions (<0.6 mm, gular to rounded (aeolian) quartz (<1 mm, mostly <0.4 mm) and
temper?), mostly composed of abundant, angular to rounded (aeo- sparce Ca-microfossils, feldspars and limonitic fragments.
lian) quartz and subordinate Ca-microfossils (foraminifera). Date: third to fourth century?
Date: late fifth century to beginning of the sixth century Origin: Africa undetermined
Origin: Northern Hammamet Gulf, probably the region of Sidi Jdidi
Africana II A?
Tripolitanian III?
112.
109. Name: Rim
Name: Rim No.: 455 (172) - 9255
No.: 1240 (188) Context: Room A, 111
Context: Room 101 Dim: D = 14.7 cm, V = 2.1 cm, HD = 3.4 cm, H = 4.6 cm
Dim: D = 14 cm, V = 1.3 cm, HD = 3.5 cm, H = 4.9 cm Colour: reddish brown (2.5YR 5/4)
Colour: red (2.5YR 4/6) VC: Hard, fine-grained fabric with frequent limestone or white reaction
VC: Hard, coarse fabric, with quartz inclusions. There are rare voids rims surrounding voids that once contained this material.
(1 mm) and very fine limestone inclusions. Date: end of the second century to beginning of the third century
Comment: not typical Tripolitania Origin: Salakta

Africana I A/B Africana II B, Pseudo-Tripolitanian?

110. 113.
Name: Rim, neck, handle Name: Rim
No.: 732 (88) No.: 140 (177) - 25965
Context: ? Context: Courtyard, 106
Dim: D = 13.3 cm, V = 2.1 cm, HD = 4.2 cm, S = 1.8 × 2.9 cm, H = 8.3 cm Dim: D = 15 cm, V = 2 cm, HD = 3.6 cm, H = 6.3 cm
Colour: red (2.5YR 5/8), SF: reddish yellow (7.5YR 7/6) Colour: brown (7.5YR 5.2)
VC: Hard, medium-grained sandy fabric, with visible quartz and moder- VC: Hard, fine-grained fabric, rare medium-sized quartz and limestone
ate limestone inclusions and microfossils. inclusions.
Thin-section: Fe-rich clay matrix. Abundant well sorted sandy inclusions Date: third century?
(temper, <0.6 mm, mostly 0.1-0.3 mm), mostly composed of angular Origin: Africa undetermined
to rounded quartz and Ca-microfossils.
Date: end of the second century to beginning of the third century
Origin: Salakta

BEZECZKY_Text.indd 91 18.03.19 14:39

 92 THE AMPHORAE

Africana II C3 Colour: red (2.5YR 4/8)

VC: Hard, fine-grained sandy fabric with quartz and rare limestone inclu-
114. sions.
Name: Rim Date: third century?
No.: 1375 (173) Origin: Africa undetermined
Context: Room 231
Dim: D = 12 cm, V = 2 cm, H = 5.5 cm 119.
Colour: red (2.5YR 5/8) Name: Rim, neck, handle
VC: Hard, fine-grained sandy fabric, with quartz and limestone inclusions No.: 59 (176)
and microfossils. Context: Room D
Date: end of the third century to first half of the fourth century Dim: D = 13 cm, V = 1.6 cm, HD = 3 cm, H = 15.3 cm
Origin: Nabeul Colour: reddish yellow (5YR 6/6)
VC: Hard, fine-grained sandy fabric
Africana II C Date: ?
Origin: Africa undetermined (uncertain typology)
115.
Name: Base Africana III A
No.: 1211 (59) - 9304
Context: ? 120.
Dim: DF = 5.6 cm, H = 27 cm Name: Rim, neck, handle
Colour: red (10R 5/8) No.: 1539 (179)
VC: Hard, fine-grained sandy fabric, with quartz and limestone inclusions Context: Dolia 5
and microfossils. Dim: D = 12.7 cm, V = 1.8 cm, HD = 3.5 cm, S = 2.1 × 2.9 cm, H =
Date: mid- to late third century 16.5 cm
Origin: Nabeul Colour: red (2.5YR5/6), SF: reddish brown (5YR 5/3) coated light brown
(7.5YR 6/4)
Africana II D VC: Hard, medium and fine-grained sandy fabric with quartz and lime-
stone or white reaction rims surrounding voids that once contained
116. this material. Irregularly fired.
Name: Rim, neck, handle Date: fourth century
No.: 462 (174) - 16210 Origin: Africa undetermined, Thaenae not excluded
Context: Room Z 2
Dim: D = 11.2 cm, V = 1.5 cm, HD = 3.5 cm, H = 10.3 cm 121.
Colour: red (2.5YR 4/8), SF: reddish brown (5YR 5/3) Name: Rim
VC: Hard, fine-grained fabric with occasional coarse limestone or white No.: 1506 (184)
reaction rims surrounding voids that once contained this material. Context: Room 213
Irregularly fired. Dim: D = 13 cm, V = 1.6 cm, HD = 4.1 cm, H = 5.4 cm
Date: mid- to late third century Colour: reddish brown (2.5YR 5/4)
Origin: Africa undetermined, Thaenae not excluded VC: Hard, fine-grained sandy fabric with quartz inclusions.
Date: fourth century
African II/III transitional Origin: Africa undetermined

117. 122.
Name: Rim Name: Base
No.: 12 (8) - 26013 No.: 34 (22) - 16164
Context: Room?, 143 Context: 36?
Colour: red (2.5YR 5/6) Dim: DF = 2.8 cm, H = 10.5 cm
VC: Hard, fine-grained fabric, with black, quartz and limestone inclu- Colour: red (2.5YR 4/8)
sions and microfossils. VC: Hard, fine-grained sandy fabric, with quartz and frequent limestone
Thin-section: Fe-rich clay matrix. Abundant, moderately sorted inclusions, or white reaction rims surrounding voids that once contained this
mostly composed of angular to rounded (aeolian) quartz (<0.4 mm), material.
Ca-microfossils and limestone fragments (<1.5 mm). Date: fourth century
Date: end of the third century to first half of the fourth century Origin: Africa undetermined, Nabeul not excluded
Origin: Africa undetermined
Africana III A/B
Africana II variant
123.
118. Name: Neck, handle
Name: Rim No.: 865 (190)
No.: 1 (3) - 26013 Context: Room A
Context: Room?, 143 Graffito: K (…), ante cocturam
Dim: DF = 5.6 cm, H = 8.7 cm Dim: S = 2 × 3 cm, H = 18 cm

BEZECZKY_Text.indd 92 18.03.19 14:39

 CHAPTER 9 – Catalogue (Bezeczky et al.) 93

Colour: red (10R 5/8) Dim: D = 13.5 cm, V = 2.9 cm, HD = 3.3 cm, H = 6.2 cm
VC: Hard, very fine-grained fabric with red, grey, white and quartz inclu- Colour: red (2.5YR 4/8)
sions. VC: Hard, fine-grained fabric with frequent fine and some coarse-sized
Date: fourth century limestone or white reaction rims surrounding voids that once con-
Origin: Africa undetermined, Nabeul not excluded tained this material and microfossils.
Thin-section: Fe-rich clay matrix. Moderately sorted inclusions with bi-
Africana III B modal distribution. Very abundant fine fraction (<0.1 mm), mainly
composed of quartz and Ca-microfossils (foraminifera). Sparce
124. coarser inclusions (<0.8 mm), composed of subrounded to rounded
Name: Rim, neck, handle (aeolian) quartz, subordinate limestone and rare fine-grained quartz-
No.: 1211 (58) - 9301 sandstone fragments.
Context: ? Date: first half of the fifth century
Dim: D = 12.6 cm, V = 2.8 cm, HD = 1.5 cm, S = 2.1 × 4.3 cm, Origin: Nabeul, zone B (probably Sidi Zahruni)
H = 9.6 cm
Colour: red (10R 576) 129.
VC: Hard, very fine-grained sandy fabric with quartz and rare limestone Name: Rim, neck, handle
inclusions. No.: 350 (72) - 9978
Date: fourth century Context: Room B2, 103
Origin: possibly Nabeul Dim: D = 10.2 cm, V = 1.9 cm, HD = 2.2 cm, H = 7.8 cm
Colour: red (2.5YR 5/8), SF: buff (10YR 8/3)
125. VC: Hard, very fine-grained sandy fabric with red, white (limestone) and
Name: Rim, neck quartz inclusions.
No.: 34 (23) - 16158 Date: mid-fifth century
Context: 36 Origin: Nabeul, zone B
Dim: D = 13 cm, V = 1.8 cm, HD = 1.5 cm, H = 6 cm
Colour: red (2.5YR 5/8) 130.
VC: Hard, fine-grained sandy fabric with quartz and rare medium-sized Name: Rim, neck, handles
limestone inclusions or white reaction rims surrounding voids that No.: 486 (185) - 10429
once contained this material. Context: ?
Date: fourth century Dim: D = 9.6 cm, V = 1.9 cm, HD = 2.2 cm, H = 8.2 cm
Origin: possibly Nabeul Colour: reddish brown (2.5 YR 4/4)
VC: Hard, coarse-grained sandy fabric with quartz and limestone inclu-
126. sions.
Name: Rim Date: fifth century?
No.: 1509 (178) Origin: Africa undetermined
Context: Room 203
Dim: D = 14.5 cm, V = 1.8 cm, HD = 3.2 cm, H = 7 cm Spatheion 1 late
Colour: red (2.5YR 5/8)
VC: Hard, fine-grained sandy fabric with quartz and limestone inclusions. 131.
Date: fourth century Name: Rim, neck
Origin: Africa undetermined (not Nabeul?) No.: 46 (37) - 15340
Context: Room Z
Africana III C Dim: D = 9.8 cm, V = 1.6 cm, HD = 2.3 cm, H = 17.5 cm
Colour: red (7.5YR 5/6), SF: very pale brown (10YR 8/3)
127. VC: Hard, very fine-grained sandy fabric with red and quartz inclusions.
Name: Rim, neck Thin-section: Fe-rich clay matrix. Well-sorted inclusions with bimodal
No.: 63 (41) - 16592 distribution. Abundant fine fraction (<0.1 mm), mainly composed
Context: 8? of quartz and Ca-microfossils (foraminifera). Relatively scarce coarser
Dim: D = 14.5 cm, V = 2.5 cm, HD = 2 cm, DN = 9.6 cm, H = 8.5 cm inclusions (<0.5 mm), composed of subangular to rounded (aeolian)
Colour: red (10R 5/8), SF: reddish yellow (7.5YR 7/4) quartz and rare fine-grained quartz-sandstone and limestone frag-
VC: Hard, fine-grained sandy fabric, with quartz and frequent limestone ments.
and rare medium-sized white reaction rims surrounding voids that Date: fifth century
once contained this material and microfossils. Origin: Nabeul, zone B
Date: end of the fourth century to first half of the fifth century
Origin: possibly Nabeul Spatheion 3A

Spatheion 1? (or Keay 35A) 132.

Name: Rim, neck, handle
128. No.: 1381 (137)
Name: Rim, neck Context: Room 61?
No.: 43 (28) - 25853 Dim: D = 7.1 cm, V = 1.5 cm, HD = 1.5 cm, H = 7.8 cm
Context: 43? Colour: red (2.5YR 5/6)

BEZECZKY_Text.indd 93 18.03.19 14:39

 94 THE AMPHORAE

VC: Hard, fine-grained sandy fabric with quartz and rare limestone inclu- 136.
sions. Name: Rim, neck
Thin-section: Fe-rich clay matrix. Very rare silty fraction (“pure” matrix). No.: 733 (83)
Abundant well-sorted sandy inclusions (0.1-0.3 mm, temper), mostly Context: Room 202
composed of angular to rounded (aeolian), rarely poly-cristalline Dim: D = 6.5 cm, V = 1.3 cm, HD = 2 cm, H = 3.2 cm
quartz. Colour: buff (10YR 7/2)
Date: late sixth century to start of the seventh century VC: Hard, fine-grained sandy fabric with red, grey and quartz inclusions,
Origin: Africa undetermined (Algeria not excluded) limestone or voids that once contained this material.
Thin-section: Ca-rich clay matrix. Moderately sorted inclusions. Very
Spatheion 3B abundant finer fraction (<0.2 mm), mainly composed of angular
quartz, Ca-microfossils and limestone fragments. Relatively sparce
133. coarser fraction (<0.6 mm), composed of subangular to rounded
Name: Rim (aeolian) quartz grains and rare limestone fragments.
No.: 680 (86) Date: seventh century
Context: ? Origin: Africa (Nabeul not completely excluded)
Dim: D = 6.7 cm, V = 0.8 cm, HD = 1.8 cm, H = 2.9 cm
Colour: yellowish red (5YR 5/8) Keay 3/5 or 64
VC: Hard, fine-grained sandy fabric with red, grey and quartz inclusions.
Thin-section: Fe, Ca-rich clay matrix. Relatively abundant, moderately 137.
sorted inclusions (<0.4 mm), mostly composed of angular to round- Name: Rim
ed quartz grains, limestone fragments and Ca-microfossils (fora­ No.: 110 (62) - 15992
mini­fera). Context: ?
Date: seventh century Dim: D = 12.5 cm, V = 2.4 cm, HD = 2.5 cm, H = 4.1 cm
Origin: Africa undetermined Colour: red (10R 5/8)
VC: Hard, fine-grained fabric with quartz and frequent limestone or
Spatheion 3C white reaction rims surrounding voids that once contained this mate-
rial.
134. Thin-section: Fe-rich clay matrix. Abundant moderately sorted inclusions
Name: Rim, neck, handles (<0.5 mm), mainly composed of angular to rounded (aeolian) quartz
No.: 907 (92) and Ca-microfossils (<0.2 mm). Abundant silty fraction.
Context: Room Z, 433? Date: fifth century
Dim: D = 6.7 cm, V = 1.4 cm, HD = 2.5 cm, H = 8.2 cm Origin: Nabeul zone B (not Sidi Zahruni)
Colour: buff (7.5YR 8/4 - 7/4)
VC: Hard, fine-grained sandy fabric with red, grey and quartz inclusions, Keay 8B
limestone or voids that once contained this material.
Thin-section: Ca-rich clay matrix. Moderately abundant, well-sorted in- 138.
clusions with bimodal distribution. Finer fraction (<0.2 mm) mostly Name: Rim, neck
composed of angular quartz, Ca-microfossils, limestone fragments No.: 1 (2) - 16292
and subordinate Fe-oxides. Coarser fraction (<0.8 mm) composed Context: Room Beta, 104
of subangular to rounded (aeolian) quartz grains and rare limestone Dim: D = 15 cm, V = 2.3 cm, HD = 3 cm, H = 6.2 cm
fragments. Colour: light red (2.5YR 6/6)
Date: seventh century VC: Hard, very fine-grained sandy fabric with frequent fine limestone or
Origin: Africa, possibly Nabeul white reaction rims surrounding voids that once contained this mate-
rial and some medium-sized quartz.
135. Date: mid- to late fifth century
Name: Rim, neck Origin: South Byzacena
No.: 635 (84)
Context: ? Keay 59 and Keay 8B / Bonifay 37-38-39
Dim: D = 7 cm, V = 1.3 cm, HD = 2.3 cm, DN = 4.8 cm, H = 9.3 cm
Colour: buff (10YR 7/2) 139.
VC: Hard, fine-grained sandy fabric with dark grey and quartz inclusions, Name: Base
limestone or voids that once contained this material. No.: 350 (71) - 9965
Thin-section: Ca-rich clay matrix. Moderately abundant, moderately sort- Context: Room B2
ed inclusions. Finer fraction (<0.2 mm) mainly composed of angular Dim: DF = 4.8 cm, H = 11.1 cm
quartz, Ca-microfossils, limestone fragments and subordinate Fe- Colour: red (2.5YR 5/8)
oxides. Coarser fraction composed of subangular to rounded (aeolian) VC: Hard, very fine-grained sandy fabric with red, dark grey and quartz
quartz grains (<0.6 mm) and rare limestone fragments (<1 mm). inclusions. Frequent fine limestone or white reaction rims surround-
Date: seventh century ing voids that once contained this material.
Origin: Africa (Nabeul not excluded) Date: fifth century +?
Origin: South Byzacena

BEZECZKY_Text.indd 94 18.03.19 14:39

 CHAPTER 9 – Catalogue (Bezeczky et al.) 95

Keay 11B var. Keay 1984, Fig. 172.2 Date: fifth century
Origin: Africa undetermined
140.
Name: Rim Keay 55
No.: 629 (79)
Context: Room 213 144.
Dim: D = 14 cm, V = 1.5 cm, HD = 1.8 cm, H = 4.6 cm Name: Rim
Colour: light reddish brown - light red (2.5YR 6/4- 6/6) No.: 38 (24) - 2124
VC: Hard, fine-grained reddish and limestone inclusions. Context: Room D
Thin-section: Fe-rich clay matrix. Relatively sparce, moderately sorted in- Dim: D = 13 cm, V = 1.9 cm, H = 12.3 cm
clusions, mostly composed of subangular to rounded (aeolian) quartz Colour: red (2.5YR 4/8), SF: buff (7.5YR 8/2)
(<0.7 mm) and subordinate Ca-microfossils (mostly <0.2 mm), with VC: Hard, very fine-grained sandy fabric with quartz inclusions, lime-
rare shale fragments and Fe-rich clay pellets (<0.5 mm). stone or voids that once contained this material.
Date: late fifth century? Date: end of the fifth century to first half of the sixth century
Origin: Africa undetermined Origin: Nabeul, zone B

Keay 35A 145.

Name: Rim, handle
141. No.: 44 (34) - 25885
Name: Rim, neck Context: 31?
No.: 108 (50) Dim: D = 12.3 cm, V = 1.9 cm, H = 11.1 cm
Context: Garden Colour: red (2.5YR 4/6), SF: light reddish brown (5YR 6/3)
Dim: D = 15 cm, V = 3.5 cm, HD = 3 cm, H = 5 cm VC: Hard, very fine-grained sandy fabric with quartz inclusions, lime-
Colour: red (2.5YR 5/8), SF: buff (7.5YR 7/4) stone or voids that once contained this material, rarely grog visible
VC: Hard, fine-grained sandy fabric with red, grey and quartz inclusions, (2 mm).
limestone or voids that once contained this material. Date: end of the fifth century to first half of the sixth century
Thin-section: Fe-rich clay matrix. Moderately abundant, moderately sorted Origin: Nabeul, zone B
inclusions with bimodal distribution. Finer fraction (<0.15 mm, in-
cluding frequent silty inclusions) composed of angular quartz, Ca- Keay 57
microfossils and rare feldspar and mica. Coarser fraction composed
of subrounded to rounded (aeolian) quartz grains (<0.5 mm) and rare 146.
fine-grained quartz-sandstone fragments (<1.5 mm). Name: Rim, neck, handle
Date: fifth century No.: 56 (38) - 23442
Origin: Nabeul? (if Nabeul, zone B) Context: 125?
Dim: D = 12 cm, V = 2.2 cm, HD = 4.1 cm, S = 2.2 × 3.6 cm, H = 16 cm
Keay 35B Colour: red (10R 5/6)
VC: Hard, very fine-grained sandy fabric with quartz and limestone inclu-
142. sions.
Name: Base Thin-section: Fe-rich clay matrix with rare inclusion-free bands. Moderate-
No.: 1230 (99) ly sorted inclusions composed of very abundant, angular to rounded
Context: Room LJ, close to the sarcophagus (aeolian) quartz (<0.4 mm), subordinate Ca-microfossils, feldspar,
Dim: DF = 6.2 cm, H = 9.2 cm mica (<0.2 mm), and rare limestone and fine-grained quartz-sand-
Colour: red (10R 5/8), SF: light red (2.5YR 6/8) stone fragments (<1.2 mm).
VC: Hard, fine-grained sandy fabric with red and quartz inclusions, lime- Date: late fifth century
stone or voids that once contained this material and microfossils. Origin: Africa undetermined (Nabeul not excluded)
Date: fifth century
Origin: Nabeul, zone B 147.
Name: Rim, neck, handle
Keay 36 No.: 44 (31) - 25872
Context: 31?
143. Dim: D = 14 cm, V = 2.3 cm, HD = 4 cm, H = 7.8 cm
Name: Rim Colour: red (10R 5/6)
No.: 668 (76) VC: Hard, very fine-grained sandy fabric with quartz inclusions, lime-
Context: ? stone or voids that once contained this material and microfossils.
Dim: D = 13 cm, V = 2.3 cm, H = 4.5 cm Thin-section: Fe-rich clay matrix. Abundant moderately sorted inclusions,
Colour: light red (10R 6/8), SF: buff (7.5YR 8/3) composed of angular to rounded (aeolian) quartz (<0.7 mm, mostly
VC: Hard, fine-grained sandy fabric with red, quartz inclusions, limestone <0.3 mm), subordinate Ca-microfossils and rare feldspar and mica
or voids that once contained this material. (<0.2 mm). Abundant silty fraction.
Thin-section: Fe, Ca-rich clay matrix with rare Fe-rich clay bands. Moder- Date: late fifth century
ately abundant, moderately sorted inclusions (<0.5 mm) composed of Origin: Africa, Nabeul?
angular to rounded (aeolian) quartz and subordinate Ca-microfossils,
limestone fragments and Fe-oxide aggregates.

BEZECZKY_Text.indd 95 18.03.19 14:39

 96 THE AMPHORAE

148. Keay 62
Name: Rim, neck, handle
No.: 44 (30) - 25873 153.
Context: 31? Name: Rim
Dim: D = 13 cm, V = 2.6 cm, HD = 4.5 cm, H = 7.4 cm No.: 140 (53) - 25945
Colour: red (10R 5/6) Context: Courtyard, 106
VC: Hard, very fine-grained sandy fabric with quartz inclusions, lime- Dim: D = 16 cm, V = 2.3 cm, HD = 2.7 cm, H = 5.4 cm
stone or voids that once contained this material. Colour: light red (10R 6/8), SF: buff (7.5YR 8/4)
Date: late fifth century VC: Hard, fine-grained sandy fabric with red, grey and quartz inclusions,
Origin: Nabeul, zone B limestone or voids that once contained this material.
Thin-section: Fe, Ca-rich clay matrix. Abundant, moderately sorted inclu-
Keay 57-56-55 sions (<0.7 mm), mostly composed of subangular to rounded (aeo-
lian) quartz, Ca-fossils (foraminifera, echinoid radioles, mollusc frag-
149. ments), and limestone fragments.
Name: Base Date: first half of the sixth century
No.: 1272 (186) Origin: possibly Moknine
Context: 29?
Dim: DF = 4 cm, H = 6.6 cm 154.
Colour: red (10R 5/8) Name: Base
VC: Hard, very fine-grained sandy fabric with quartz inclusions, lime- No.: 1240 (112)
stone or voids that once contained this material. Context: Room 101
Date: late fifth century to first half of the sixth century Dim: DF = 7 cm, H = 11.2 cm
Origin: Nabeul, zone B Colour: reddish grey (10R 5/1)
VC: Hard, fine-grained fabric with frequent limestone or white reaction
Keay 62 Q or Albenga 11/12 rims surrounding voids that once contained this material.
Date: first half of the sixth century
150. Origin: possibly Nabeul
Name: Rim
No.: 44 (32) - 25863 (Keay 62 or 61?)
Context: 31?
Dim: D = 14 cm, V = 1.7 cm, HD = 5 cm, H = 5.5 cm 155.
Colour: red (10R 5/6), SF: buff (10YR 8/4) Name: Base
VC: Hard, very fine-grained sandy fabric with quartz inclusions and voids. No.: 1240 (114)
Thin-section: Fe, Ca-rich clay matrix. Abundant, moderately sorted inclu- Context: Room 101
sions, mainly composed of subangular to rounded (aeolian) quartz Dim: DF = 6.8 cm, H = 7.8 cm
(<0.7 mm, mostly <0.4 mm) and Ca-microfossils (<0.2 mm). Rela- Colour: grey (2.5YR 5/0)
tively sparce silty fraction. VC: Hard, fine-grained fabric with frequent limestone or white reaction
Date: end of the fifth century to first half of the sixth century rims surrounding voids that once contained this material.
Origin: Africa undetermined Date: sixth or seventh century
Origin: Africa (Nabeul not excluded)
151.
Name: Rim, neck Keay 61
No.: 43 (29) - 25850
Context: 43? 156.
Dim: D = 15 cm, V = 2 cm, HD = 5.2 cm, H = 7.9 cm Name: Base
Colour: pink (5YR 7/4) No.: 207 (75) - 2344
VC: Hard, very fine-grained sandy fabric common with quartz inclusions. Context: Room T
Thin-section: Ca-rich (?) altered clay matrix. Moderately sorted inclusions Dim: DF = 6.3 cm, H = 6.1 cm
(<0.6), mainly composed of subangular to rounded (aeolian) quartz Colour: light red (2.5YR 6/6)
and subordinate limestone fragments and Ca-microfossils. VC: Hard, fine-grained sandy fabric with red, grey and quartz inclusions,
Date: end of the fifth century to first half of the sixth century limestone or voids that once contained this material and microfossils.
Origin: Africa undetermined Date: seventh century
Origin: Moknine probable
152.
Name: Base 157.
No.: 70 (43) - 10405 Name: Base
Context: 69? No.: 203 (189) - 24687
Dim: DF = 4.1 cm, H = 10.8 cm Context: Garden
Colour: red (2.5YR 4/8) Dim: DF = 5 cm, H = 5.7 cm
VC: Hard, very fine-grained sandy fabric with quartz inclusions and voids. Colour: red (2.5YR 5/8)
Date: end of the fifth century to first half of the sixth century VC: Hard, fine-grained sandy fabric with red, grey and quartz inclusions,
Origin: Africa undetermined limestone or voids that once contained this material and microfossils.
Date: seventh century
Origin: Moknine probable

BEZECZKY_Text.indd 96 18.03.19 14:39

 CHAPTER 9 – Catalogue (Bezeczky et al.) 97

Keay 61A/D Dim: D = 13.5 cm, V = 1.5 cm, HD = 5 cm, H = 6.1 cm

Colour: red (2.5YR 5/8)
158. VC: Hard, medium-grained sandy fabric with red and quartz inclusions,
Name: Rim rare coarse limestone specks.
No.: 1381 (183) Date: mid- to late seventh century
Context: Room 61? Origin: Africa undetermined (Sahel region not excluded)
Dim: D = 15.3 cm, V = 2 cm, HD = 4.6 cm, H = 5.8 cm
Colour: light red (2.5YR 6/8) 163.
VC: Hard, fine-grained sandy fabric with red, grey and quartz inclusions, Name: Base
limestone or voids that once contained this material and microfossils. No.: 1522 (187)
The external surface has a white or cream colour. Context: Room 7A
Date: seventh century Dim: DF = 5.7 cm, H = 8.5 cm
Origin: Moknine probable Colour: red (10R 5/8)
VC: Hard, medium-grained sandy fabric with red and quartz inclusions
159. and voids, occasionally grog visible.
Name: Base Date: mid- to late seventh century?
No.: 12 (7) - 25994 Origin: Africa undetermined (Sahel region not excluded)
Context: Room ?, 143
Dim: DF = 6.2 cm, H = 9.3 cm Keay 34
Colour: light red (2.5YR 6/8), SF: very pale brown (10YR 7/3)
VC: Hard, medium-grained sandy fabric with red, grey and quartz inclu- 164.
sions, limestone or voids that once contained this material and micro- Name: Rim, neck
fossils. The external surface has a white or cream colour. No.: 52 (36) - 26487
Date: seventh century Context: Room 12
Origin: possibly Moknine Dim: D = 13.5 cm, V = 2.3 cm, HD = 3.4 cm, H = 5.6 cm
Colour: red (10R 4/6), SF: weak red (2.5YR 5/2)
Keay 61? VC: Hard, fine-grained sandy fabric with quartz inclusions, limestone or
voids that once contained this material and microfossils.
160. Thin-section: Fe-rich (vitrified) clay matrix. Abundant, moderately sorted
Name: Base inclusions (<0.5 mm), mainly composed of angular to rounded (aeo-
No.: 12 (35) lian) quartz, subordinate Ca-microfossils, limestone fragments (com-
Context: Garden, 143? pletely dissociated), and rare feldspar and Fe-rich clay nodules.
Dim: DF = 6.8 cm, H = 10.5 cm Date: sixth to seventh century
Colour: dark reddish grey (5YR 4/2), SF: yellowish red (5YR 5/6) Origin: Africa undetermined (Rougga not excluded, high-T variant)
VC: Hard, medium-grained sandy fabric common with quartz and rare
limestone inclusions. 165.
Thin-section: Ca-rich clay matrix. Moderately abundant, moderately sort- Name: Rim, neck
ed inclusions (<0.6 mm), mainly composed of angular to rounded No.: 750 (77)
(aeolian) quartz and subordinate Ca-microfossils. Context: ?
Date: seventh century? Dim: D = 15.8 cm, V = 2.5 cm, HD = 3.5 cm, H = 5.4 cm
Origin: Africa undetermined Colour: red (2.5YR 5/8)
VC: Hard, very fine-grained sandy fabric with quartz inclusions, lime-
Keay 61C stone or voids that once contained this material and microfossils.
Thin-section: Fe-rich clay matrix. Well-sorted inclusions with bimodal
161. distribution. Abundant finer fraction (<0.1 mm) mainly composed
Name: Rim, neck of angular quartz and subordinate Ca-microfossils. Moderately abun-
No.: 935 (96) dant coarser fraction (<0.5 mm), essentially composed of subangular
Context: Room Z to rounded (aeolian) quartz.
Dim: D = 15 cm, V = 1.8 cm, HD = 3.5 cm, H = 8.2 cm Date: sixth to seventh century
Colour: yellowish red (5YR 5/6) Origin: region of Rougga (typical fabric)
VC: Hard, medium-grained sandy fabric common with frequent lime-
stone or white reaction rims surrounding voids that once contained 166.
this material. The void size is occasionally 2-3 mm. Name: Base
Date: end of the sixth century to first half of the seventh century No.: 31 (17) - 1467
Origin: Henchir Chekaf probable Context: Old street
Dim: DF = 6.5 cm, H = 6.2 cm
Keay 8A / Bonifay 50 Colour: red (10R5/6)
VC: Hard, fine-grained sandy fabric with quartz inclusions, limestone or
162. voids that once contained this material and microfossils.
Name: Rim Thin section: Fe-rich clay matrix. Abundant, moderately sorted inclusions
No.: 1522 (175) (<0.5 mm), mainly composed of subangular to rounded (aeolian)
Context: Room 7A quartz and scarce (or poorly visible) Ca-microfossils.

BEZECZKY_Text.indd 97 18.03.19 14:39

 98 THE AMPHORAE

Date: sixth to seventh century Origin: Lusitania

Origin: possibly region of Rougga Comment: Keay 22, small Lusitanian version Almagro 50.

Keay 1B? 172.

Name: Neck
167. No.: 1356 (195)
Name: Rim, neck, handle Context: Room 112
No.: 455 (65) - 9246 Stamp: [---]VR
Context: Room A, 111 Dim: DF = 6.8 cm, H = 7.8 cm
Dim: D = 14 cm, S = 2.2 × 3.7 cm, H = 14.2 cm Colour: reddish yellow (5YR 7/6 - 7/8)
Colour: Irregularly fired, light red (10R 6/8) and buff (10YR 8/3) VC: Soft, smooth fabric. Rare red, black and limestone inclusions. It has
VC: Hard, fine-grained sandy fabric common with red, grey and quartz a lot of small silver mica inclusions. The fabric is from Marseille or a
inclusions, limestone or voids that once contained this material. nearby place in the South Narbonensis.
Thin-section: Fe-rich clay matrix. Moderately abundant, moderately sorted Date: Augustan to Julio-Claudian period
inclusions (<0.7 mm) with bimodal distribution, mainly composed of Comment: Marseille, it is a Gauloise 2 type. The stamp is similar to other
subangular to rounded (aeolian) quartz, subordinate Ca-microfossils from Gauloise 2 produced in Marseille.
(<0.2 mm), sparce calcite and limestone fragments and occasional
cherts and clinopyroxene.
Date: fourth century? 173.
Origin: Africa undetermined. Very different fabric from the typical Keay Name: Rim, neck, handle
1B ones. No.: 1403 (145)
Context: Room T
Storage jars? Dim: D = 12.2 cm, V = 1 cm, H = 5.4 cm
Colour: reddish yellow (7.5 YR 8/6)
168. VC: Hard, fine fabric. There are often scattered red and grey inclusions
Name: Base and with mica and small voids.
No.: 1529 (182) Date: fifth to start of the sixth century A.D.
Context: Room 203 Origin: Icaria?
Dim: DF = 7.6 cm, H = 11 cm Comment: Adamschek 22/Torone VII small version
Colour: reddish brown (5YR 5/2)
VC: Hard, sandy fabric. 174.
Date: ? Name: Rim, neck, handle
Origin: Africa undetermined No.: 1385 (144)
Context: ?
169. Dim: D = 11.2 cm, V = 1.3 cm, S = 2.2 × 3.8 cm, H = 8.5 cm
Name: Base Colour: reddish yellow (5YR 7/8-6/8)
No.: 59 (39) - 9285 VC: Hard and very coarse, with big and medium transparent rounded in-
Context: Room D, 201 clusions. Irregularly fired.
Dim: DF = 8.6 cm, H = 2.2 cm Date: fifth to seventh century A.D.
Colour: red (2.5YR 5/6) Origin: Icaria?
VC: Hard, sandy fabric, microfossils. Comment: Adamschek 22/Torone VII
Date: ?
Origin: Africa undetermined 175.
Name: Rim, neck, handle
Miscellaneous No.: 1321 (140)
Context: Room A / XA
170. Dim: D = 9.8 cm, V = 0.9 cm, S = 1.9 × 3.3 cm, H = 5.4 cm
Name: Rim, neck, handles Colour: reddish yellow (7.5YR 6/8)
No.: 175 (171) - 16240 VC: Hard and coarse fabric. It has a large amount small and medium mica
Context: Room E inclusions and, in smaller amounts, some angular transparent quartz,
Dim: D = 7.3 cm, V = 0.9 cm, HD = 1.4 cm, S = 2.1 × 3.2 cm, H = 14 cm lime and occasional small red inclusions.
Colour: red (2.5YR 5/8) Origin: probably Asia Minor production
VC: Hard, very fine, smooth fabric. Rare grey inclusions and voids.
176.
171. Name: Complete rim and neck and complete handle
Name: Base No.: 453 (198)
No.: 1344 (132) Context: Room A
Context: Room 204 Dim: D = 8.8 cm, DN = 6.4 cm, S = 2.5 × 3.8 cm, H = 13 cm
Dim: D = 7.8 cm, V = 1.2 cm, H = 9.6 cm Colour: light red (2.5YR 6/8)
Colour: red (2.5YR 5/8). VC: Very hard, very fine sandy fabric. Moderate red and grey inclusions.
VC: Very hard, coarse Lusitanian fabric, with a sparce scatter of red, black Date: seventh century A.D.
and limestone inclusions. Oxidized fired. Coated with grey slip. Comment: Late version of LRA 1, Aegean production, maybe from Cos,
Date: fourth to sixth century A.D. but without mica.

BEZECZKY_Text.indd 98 18.03.19 14:39

 CHAPTER 9 – Catalogue (Bezeczky et al.) 99

177. 181.
Name: Rim, neck Name: Neck and one complete handle
No.: 1297 (138) No.: 386 (197) - 25126
Context: Room 201 Context: Room GL
Dim: D = 9.7 cm, S = 1.9 × 3.4 cm, H = 8.4 cm Dim: DN = 5.5 cm, S = 1.5 × 2.5 cm, H = 13 cm
Colour: dark red or red (10R 3/6 - 4/6) Colour: red (10R 5/8).
VC: Very hard and coarse fabric. Common small or medium colourless, VC: Hard, fine fabric. Often scattered white limestone fragments, rare red
grey and black grains and the mica is abundant. and grey inclusions.
Date: fifth to seventh century A.D. Date: fifth to seventh century A.D.
Origin: Northern Sicily
Comment: Crypta Balbi 2 182.
Name: Body frgm.
178. No.: 846 (199)
Name: Rim Context: Room G
No.: 1280-2 (146) Titulus pictus: red paint: K(…)
Context: Room 103A Dim: H = 11 cm
Dim: DF = 6.8 cm, H = 7.8 cm Colour: red (2.5YR 5/8)
Colour: reddish yellow (5YR 6/8) VC: Hard, fine fabric. There are often scattered white limestone frag-
VC: Soft, smooth fabric. Common very fine limestone inclusions, moder- ments, the microfossils are common.
ate microfossils.
Date: fifth to seventh century A.D. 183.
Origin: Greece Name: Body frgm.
Comment: TRC 2 (Portale – Romeo 2000). Late Cretan Amphora. No.: 734 (200)
Context: ?
179. Graffito: Υε(…)
Name: Rim, neck and beginning of both handles Dim: H = 11 cm
No.: 174 (194) Colour: red (10YR 5/8)
Context: Room 101 VC: Hard, coarse, micaceous Ephesian fabric.
Dim: D = 5.7 cm, DN = 4.6 cm, S = 1.9 × 4.2 cm, H = 7.2 cm Date: fifth to seventh century A.D.
Colour: reddish yellow (7.5YR 7/6). Origin: Ephesus region
VC: Hard, coarse fabric. Large and medium limestone and quartz inclu- Comment: LR 3 or Ephesus 56 type
sions, some long voids and occasional flakes of mica.
Date : fifth to seventh century A.D. 184.
Origin: Crete Name: Body frgm.
Comment: TRC 1 or TRC 2 No.: 38 (26) - 2130
Context: Room D
180. Graffito: IVLIVSSIFA. The problem is how to cut the words: Iuli Ussi
Name: Body, one handle and almost complete neck Fa[---] ? or Iulius Sifa[---] ? Perhaps a modern carving?
No.: 1538 (166) Dim: V = 1.1-1.4 cm, H = 10.5 cm
Context: 27? Colour: pale red (10R 6/4)
Dim: DN = 6 cm, S = 1.6 × 2.3 cm, H = 18,3 cm VC: Hard, fine fabric, red and grey inclusions, microfossils.
Colour: buff (5YR 8/2 - 7/2)
VC: Hard, fine fabric, scattered red, grey and white inclusions and some
coarse limestone grains.
Date: fifth to seventh century A.D.
Origin: Crete
Comment: TRC 4

BEZECZKY_Text.indd 99 18.03.19 14:39

BEZECZKY_Text.indd 100 18.03.19 14:39
 PLATE 1

7
1

10
6

Lamboglia 2 (nos. 1-3), Dressel 6A (nos. 4-8), ante Dressel 6B (nos. 9-10) amphorae.

BEZECZKY_Text.indd 1 18.03.19 14:39

 PLATE 2

18

11

19

20
12

21

13

22

14

15 23

16

17 24

Dressel 6B (nos. 11-24) amphorae.

BEZECZKY_Text.indd 2 18.03.19 14:39

 PLATE 3

25
29

26
30

31
27

32

28 33

Dressel 6B (no. 25), Fažana 1 (no. 26), Fažana 2 (no. 27), Porto Recanati (nos. 28-29), Forlimpopoli (nos. 30-31),
Aquincum 78 (nos. 32-33) amphorae.

BEZECZKY_Text.indd 3 18.03.19 14:39

 PLATE 4

39

34

40

35
42

43

36

44

37

45

38
47

46 48

Dressel 2-4 (nos. 34-36), Dressel 7-11 (no. 37), Beltran IIA (no. 38), Dressel 14 (no. 39), Dressel 20 (nos. 40, 42-45),
Almagro 50A/Key XXII (no. 46), Almagro 51A-B (nos. 47-48) amphorae.

BEZECZKY_Text.indd 4 18.03.19 14:39

 PLATE 5

41

49

51 50

Dressel 20 (no. 41), Almagro 51C (no. 49), Beltran 68 (nos. 50-51) amphorae.

BEZECZKY_Text.indd 5 18.03.19 14:39

 PLATE 6

52

58

53

60

54

61

55

62

56

63

57

59 64

Gauloise 4 (no. 52), Mid Roman 1 (nos. 53-54), Palatine East 1/LRA 1 (nos. 55-56), Crete TRC 2 (no. 57),
Rhodian (nos. 58-60), Knidian (no. 61), Koan with pinched handle (no. 62), Koan (nos. 63-64) amphorae.

BEZECZKY_Text.indd 6 18.03.19 14:39

 PLATE 7

65

66

67

68
72

69
76

70

77

71

73

78
74 75

Koan (no. 65), Dressel 5 (no. 66), Crétoise AC 4 (nos. 67-69), Agora G 199/Pinched handle (no. 70), Agora F 66 (no. 71),
Agora M 125 (no. 72), Agora M 279 similis (no. 73), Agora M 126 (nos. 74-76), Agora M 240 (no. 77),
Kapitän I (no. 78) amphorae.

BEZECZKY_Text.indd 7 18.03.19 14:39

 PLATE 8

79 85

86

80

81 87

88

82
89

83 90

84 91

Kapitän II (no. 79), Agora M 273 (no. 80), Late Roman Amphora 1 (nos. 81-86), Late Roman Amphora 2
(nos. 87-88), Late Roman Amphora 3 (nos. 89-90), Ephesus 56 (no. 91) amphorae.

BEZECZKY_Text.indd 8 18.03.19 14:39

 PLATE 9

92 99

93
100
94

95

101

102

96

103

97

104
98

Ephesus 56 (no. 92), Agora M 307 (nos. 93-96), Late Roman Amphora / Gaza (nos. 97-99), Late Roman Amphora 13
(no. 100), Samos Cistern (no. 101), Beirut (no. 102), Cretan AC1C / MRC 3 (no. 103), M 235/6 (no. 104) amphorae.

BEZECZKY_Text.indd 9 18.03.19 14:39

 PLATE 10

112
105

113

106
114

107

108

115
109

116

110

117

111 118

Late Roman Amphora 7 (no. 105), Sinope C (nos. 106-107), Hammamet 3 (no. 108), Tripolitanian III?
(no. 109), Africana I A/B (no. 110), Africana I variant? (no. 111), Africana II A? (no. 112), Africana II B, Pseudo-
Tripolitanian? (no. 113), Africana II C3 (no. 114), Africana II C (no. 115), Africana II D (no. 116),
Africana II/III transitional (no. 117), Africana II variant (no. 118) amphorae.

BEZECZKY_Text.indd 10 18.03.19 14:39

 PLATE 11

125

119

126

127

120

128
121

122 129

130

123

124 131

Africana II variant (no. 119), Africana III A (nos. 120-122), Africana III A/B (no. 123), Africana III B
(nos. 124-126), Africana III C (no. 127), Spatheion 1? or Keay 35A (no. 128), Spatheion 1 (nos. 129-130),
Spatheion 1 late (no. 131) amphorae.

BEZECZKY_Text.indd 11 18.03.19 14:39

 PLATE 12

141

132

133

142

134 143

135

144

136

137
145

138

146

139

140 147

Spatheion 3A (no. 132), Spatheion 3B (no. 133), Spatheion 3C (nos. 134-136), Keay 3/5 or 64 (no. 137), Keay 8B (no. 138),
Keay 59 and Keay 8B (no. 139), Keay 11B var. Keay 1984, Fig. 172.2 (no. 140), Keay 35A (no. 141), Keay 35B (no. 142), Keay 36
(no. 143), Keay 55 (nos. 144-145), Keay 57 (nos. 146-147) amphorae.

BEZECZKY_Text.indd 12 18.03.19 14:39

 PLATE 13

156
148

157

149

158

150

159

151

160

152

153
161

162
154

155 163

Keay 57 (no. 148), Keay 57-56-55 (no. 149), Keay 62 Q or Albenga 11/12 (nos. 150-152), Keay 62 (nos. 153-154),
Keay 62 or 61? (no. 155), Keay 61 (nos. 156-157), Keay 61A/D (nos. 158-159), Keay 61? (no. 160), Keay 61C (no. 161),
Keay 8A (nos. 162-163) amphorae.

BEZECZKY_Text.indd 13 18.03.19 14:39

 PLATE 14

164

165 170

173
166

174
167

175

168

176

169
177

178
171

Keay 34 (nos. 164-166), Keay 1B ? (no. 167), Storage jars? (nos. 168-169), Miscellanous (nos. 170-171, 173-178) amphorae.

BEZECZKY_Text.indd 14 18.03.19 14:39

 PLATE 15

179

181

180

172 182

183

184

Miscellanous (nos. 172, 179-184) amphorae.

BEZECZKY_Text.indd 15 18.03.19 14:40

 PLATE 16

5 11

7 15

12
16
17

13

14 18

19 19

Lamboglia 2 (nos. 2-3), Dressel 6A (nos. 5, 7), Dressel 6B (nos. 11-19) amphora stamps.

BEZECZKY_Text.indd 16 18.03.19 14:40

 PLATE 17

20

38

21

22 40

86 100

108

123

Dressel 6B (nos. 20-22), Beltran 2A (no. 38), Dressel 20 (no. 40), Late Roman Amphora 1 (no. 86), Late Roman Amphora 13 (no. 100),
Hammamet 3A (no. 108), Africana III A/B (no. 123).

BEZECZKY_Text.indd 17 18.03.19 14:40

 Pla

PLATE 18

Dressel 6B earliest form 9 Dressel 6B 11

Dressel 6B 12 Dressel 6B 13

Dressel 6B 14 Dressel 6B 15

Dressel 6B 16 Dressel 6B 17
Photomicrographs Dressel 6B (nos. 9, 11-17).

BEZECZKY_Text.indd 18 18.03.19 14:40

 Plate 19

PLATE 19

Dressel 6B 18 Dressel 6B 19

Dressel 6B 20 Dressel 6B 21

Dressel 6B 22 Beltran 68 50

Beltran 68 51 Hammamet 3A 108

Photomicrographs Dressel 6B (nos. 18-22), Beltran 68 (nos. 50-51), Hammamet 3A (no. 108).

BEZECZKY_Text.indd 19 18.03.19 14:40

 PLATE 20

Africana I A 110 Africana I A variant? 111

Africana II/III transitional 117 Spatheion 1 128

Spatheion 1 late 131 Spatheion 3A 132

Spatheion 3B 133 Spatheion 3C 134

Photomicrographs Africana I A/B (no. 110), Africana I variant (no. 111), Africana II/III transitional (no. 117),
Spatheion 1? or Keay 35A (no. 128), Spatheion 1 late (no. 131), Spatheion 3A (no. 132), Spatheion 3B (no. 133),
Spatheion 3C (no. 134).

BEZECZKY_Text.indd 20 18.03.19 14:40

 PLATE 21

Spatheion
Spatheion3C
3C 135
135 Spatheion3C
Spatheion 3C 136
136

Keay3/5
Keay 3/5or
or64
64 137
137 Keay11B
Keay 11Bvar.
var.Keay
Keay1984,
1984,Fig.
Fig.172.2
172.2 140
140

Keay
Keay35A
35A 141
141 Keay
Keay36
36 143
143

Keay
Keay57
57 146
146 Keay
Keay57
57 147
147

Photomicrographs Spatheion 3C (nos. 135-136), Keay 3/5 or 64 (no. 137), Keay 11B var. Keay 1984, Fig. 172.2 (no. 140),
Keay 35A (no. 141), Keay 36 (no. 143), Keay 57 (nos. 146-147).

BEZECZKY_Text.indd 21 18.03.19 14:40

 PLATE 22

Keay 62 Q or Albenga 11/12 150 Keay 62 Q or Albenga 11/12 151

Keay 62 153 Keay 61? 160

Keay 34 164 Keay 34 165

Keay 34 166 Keay 1B? 167

Photomicrographs Keay 62 Q or Albenga 11/12 (nos. 150-151), Keay 62 (no. 153), Keay 61? (no. 160),
Keay 34 (nos. 164-166), Keay 1B? (no. 167).

BEZECZKY_Text.indd 22 18.03.19 14:40

 EPIGR APHY

BEZECZKY_Text.indd 123 18.03.19 14:40

BEZECZKY_Text.indd 124 18.03.19 14:40
 125

10 CALENDAR GRAFFITI ON DRESSEL 20 AMPHORAE.

ASIATICUS: ANOTHER PARADIGMATIC CASE WITH
A NEW FIND FROM BRIJUNI1
PIERO BERNI MILLET

Dressel 20 globular amphorae from Baetica usually preserve impressed with a clay signaculum on the amphora handle,
different types of inscriptions, among which the best known thus recording the batch of amphorae the potters had been
are ante cocturam graffiti, stamps, painted inscriptions (tit- commissioned to produce by the actual or temporary owner
uli picti) and post cocturam graffiti.2 All were added at dif- of the figlina.
ferent times by different hands using diverse writing tech-
niques. Their purpose had to do with each of the stages in
the life cycle of these olive oil amphorae: the manufacture
of the pottery object, its filling with olive oil, its marketing,
transportation and distribution, sale and consumption, and
finally, its possible reuse for a completely different purpose.
In the initial stage of this journey, the so-called ante coc-
turam graffiti and the stamps were applied in the Baetican
figlinae on the banks of the Guadalquivir and Genil Rivers,
where the manufacturing centres for Dressel 20 amphorae
were located.3 It was in these factories that the potters used
their fingers, a cane or any other pointed object to score the
so-called ante cocturam graffiti on different parts of the ves-
sel while the clay was still soft. The purpose was to have an
efficient means of checking their contribution to the work in
the complex organisational structure of a Roman industrial
pottery. The same workers also stamped the amphorae that
were finished and ready for firing. The stamp was normally

1
Our most sincere gratitude to the people mentioned below for
allowing us to include unpublished material of great importance for
this paper. To Joost van den Berg for the valuable set of Dutch graffiti
documented in the collections of the Radboud University Nijmegen
(Department of Roman Provincial Archaeology) Photographic
Archive. To Xavier Deru of the Université Lille 3 for very kindly
placing in our hands the photograph of the Asiaticus calendar graffito
found in Reims. To the archaeologists Guillaume Maza and Toni
Silvino for the graffito from Digoin. To Raphaël Macario for the
graffito of Béziers, and Andrew and Barbara Birley for the graffito
of Vindolanda. To Ivan Di Stefano Manzella and Diana Gorostidi
Pi for their valuable comments and suggestions. To Tamás Bezeczky
and Horacio González Cesteros for their help with the graphic Fig. 10.1 A Dressel 20 globular amphora with a stamp on the handle, tit-
documentation of the amphora with inscription from Brijuni. uli picti on the upper half of the container and an ante cocturam
2
Rodríguez Almeida 1984. graffito with the calendar date at the bottom (Berni Millet
3
Berni Millet 2008. 2008, Fig. 31).

BEZECZKY_Text.indd 125 18.03.19 14:40

 126 EPIGRAPHY

Fig. 10.2 Stages in the manufacture of the Dressel 20 amphora (Berni Millet 2008, Fig. 1).

1 THE MANUFACTURING STAGES OF A there for several weeks, as the thickness of the amphora walls
DRESSEL 20 AMPHORA meant it had to be dried slowly, carefully and safely. Once
dry, the belly was returned to the workshop, where it was
During the manufacturing phase of the pottery vessel, the
placed vertically – this time the right way up – on an earth-
different types of ante cocturam graffiti (numbers, initials,
enware pot (Fig. 10.2.3). Then the upper chamber with the
marks, names, dates, etc.) were written on various parts of
neck and the mouth was fitted to the belly (Fig. 10.2.4).
the clay object (foot, base, belly, neck, handles, etc.) in a
Before the neck was shaped and finished off with the lip,
pre-established order that followed the Dressel 20 amphora
both parts were joined by pressing the back along the seam.
production line. Therefore – as already observed by Rodrí-
Finally, the handles were added and the finished amphora
guez Almeida4 – the graffiti did not correspond to the indi-
was stamped (Fig. 10.2.5).
vidual whims of the workers, but was an integral part of a
functional recording system linked to the complex process
of manufacturing globular amphorae. Each type of graffito 2 NOMINAL GRAFFITI
belongs to a stage in the making of the object, ordered in
It was precisely during the initial phase that the most inter-
terms of the time and the different areas of the production
esting graffiti were noted on the base of the vessel, while the
line. With their annotations the members of the team of spe-
large globular body was still upside down with the so-called
cialist potters left clues as to how their collective artisanal
“breathing” hole recently sealed by the spike. These graffiti
work was organised.
were written in cursive in a relatively small script, except for
The so-called calendar graffiti were always noted next
rare exceptions when rustic capitals were used. They were
to the foot on the base of the amphora (Fig. 10.1) with the
not written by the potters, but by well-educated people with
text inverted with respect to the vertical position of the con-
a perfect command of writing who were in charge of super-
tainer. To understand their function and meaning it is neces-
vising the day-to-day work of the potters in the figlina.
sary to know how a Dressel 20 amphora was made.5 First we
The writing technique has been thoroughly analysed and
have to take into account the fact that the manufacture and
described by Rodríguez Almeida.7 In summary, this type
mounting of the Dressel 20 amphora was carried out at two
of regular, ordered and well distributed cursive inscription
different times and could take several weeks.6
was written using a stylus with a well-sharpened point. As
First the large spherical belly was worked upside down
we will see from the examples illustrated in this paper, the
with the spike pointing upwards on the turntable of a pot-
ordinatio is perfect and reveals the expert hand of a person
ter’s wheel, which was probably removable (Figs. 2.1 and
with a complete command of writing. In general the pal-
2.2). The large globular body was then taken on the sup-
aeography is quite homogenous and variations in the letters
port used on the potter’s wheel itself to the dryer. It stayed
are minimal.
Thanks to the epigraphic material from Monte Testac-
4
Rodríguez Almeida 1993, 95. cio (Rome) we are able to classify with a certain precision
5
Remesal Rodríguez 1977-78, 94; Rodríguez Almeida 1984,
162-164; Rodríguez Almeida 2000, Fig. 9
6
Berni Millet 2008, 34-36. 7
Rodríguez Almeida 1991; Rodríguez Almeida 1993.

BEZECZKY_Text.indd 126 18.03.19 14:40

 CHAPTER 10 – Calendar graffiti on Dressel 20 amphorae (Berni Millet)
Fig. 10.3 Nominal cursive graffiti: (1) single name in the genitive “Fidenti”; (2) single name in the nominative “Hermeros”; (3) pair of names in the geni-
tive “Quieti / Caprari”; (4) three simple names “Erotis / Ga[---]o / Pri( )”.
the different types of ante cocturam cursive graffiti by tak-
ing into account the established formulas, as Rodríguez
Almeida did.8
The most numerous are the nominal graffiti that con-
sist of simple names of workers that are written in full, as
separate words or paired with another simple name. The
majority of isolated nominal graffiti are written in the geni-
tive case (Fig. 10.3.1): “Antoni”, “Asini”, “Bari”, “Caprari”,
“Gallici”, “Clementini”, “Elpidori”, “Felicis”, “Fidenti”, “Fortu-
nati”, “Iuliani”, “Marini”, Musici”, “Nigrini”, “Olympici”, “Pal-
ladi”, “Primi”, “Quinti”, “Rufini”, “Secundi”, “Secuti”, “Tro-
phimi”, “Tuccudae”, “Vitalis”, etc., and less frequently in the
nominative case (Fig. 10.3.2): “Barittius”, “Capra”, “Felix”,
“Hermeros”, “Indus”, “Innoce(n)s”, “Maternus”, “Septumius”,
etc. The so-called binominal graffiti consist of two associ-
ated cognomina written in the same hand on two different
lines, with the two names normally expressed in the genitive
(Fig. 10.3.3): “Quieti / Caprari”.9 We even know of a rare case
with three simple names, the first in the genitive, the second
in the nominative, and the third abbreviated (Fig. 10.3.4):
“Erotis / Ga[---]o / Pri( )”.10
As we have seen so far, one or several of the workers who
participated in the first manufacturing phase of the large
globular body of the Dressel 20 amphora can be found in the
nominal cursive graffiti. Behind these names lies a plurality
of specialists with different responsibilities in the productive
organisation of the pottery.
Over the years Rodríguez Almeida has made a series of
observations about the possible purpose and meaning of the
cursive graffiti that merit our attention. In 1984 he speculated
as to whether the binominal cursive graffiti corresponded
8
Rodríguez Almeida 1993, 104-105.
9
Aulas 1980-81, no. 36, Pl. 7.
10
Ehmig 2003, no. 177, Pl. 50.
BEZECZKY_Text.indd 127
127
to the potter and the business associate responsible for the
operation of assembling the two parts of the amphora: “oper-
aio al tornio o alla ruota e del suo partner di assemblaggio di
collo e manici”.11 In 1993 he qualified the question with new
data and observations. He first emphasised – as a curious
detail to be taken into account – the predominance of Latin
names in the lists of cognomina, perhaps because a part of
the signatories involved in these supervision operations were
free citizens.12
Equally interesting is his suggestion that we attempt to
discern the degree of job-related responsibility of the differ-
ent people, taking into consideration the grammatical case
of the Latin declination in which the names are written.13 If
the name is declined in the second case, he says, we should
implicitly understand the noun opus, although it is uncer-
tain whether it refers to the amphora or the signature itself.
Meanwhile, with the nominative case we should implicitly
understand the verb fecit with reference to the potter. In this
way, a binominal graffito with the first name in the genitive
and the second in the nominative could be the person in
charge of amphora production (officinator) and the person
who actually made the object (figulus).
This hypothesis appears to be confirmed, at least in part,
by the flashy ante cocturam binominal graffiti from the Villar
de Brenes pottery (Brenes, Seville) for a specific phase of its
industrial activity, which is well dated at Monte Testaccio to
the late Antonine period.14 From Dressel’s prospection in the
Littera L at Monte Testaccio, dated to 179 A.D., we have the
graffiti pairs “Herme[tis] / [---]libes” (CIL, XV. 3591) and
“[Milo]nis / [S]osumus” (CIL, XV. 3608). They are also doc-
11
Rodríguez Almeida 1984, 262-264; Rodríguez Almeida
1991, 36.
12
Rodríguez Almeida 1993, 96, note 6.
13
Rodríguez Almeida 1993, 104.
14
Berni Millet 2008, 145 and 239 ff.
18.03.19 14:40
 128
Fig. 10.4 Correlations between stamps and binominal graffiti on pottery from Villar de Brenes (ca. 179 A.D.).
umented in their place of origin in Andalusia on two shards
of amphora belly: “[Her]metis / [---]++mu[s]” and “[H]
erm[etis] / [------]”.15 We can also add to this list the find of
“[Ca]llisti / [---]es[---]” during excavations at Portus Dubris
(Dover, Kent, UK) in 1951-52.16 All the above-mentioned
examples have another peculiarity in common: they are all
written in rustic capitals (Fig. 10.4).
The Villar de Brenes stamps17 from precisely this late
manufacturing period provide us with one of the keys to
understanding the function and meaning of these binomi-
nal graffiti. The stamping system at that time was based on a
group of cognomina written in the nominative together with
the verb fecit: AVGVSTALF, CALLISTVSF, HERMESF,
MILONF, ROMVLVSF.18 There can be no doubt that Cal-
listus, Milon(ius) and Hermes are the same people we find
declined in the genitive on the first line of the ante cocturam
graffiti, while the other person expressed in the nominative
has never before been seen on the Villar de Brenes amphora
stamps.
How are we to interpret this curious coincidence of names
in two contemporary records of different types? Perhaps the
presence of Callistus, Hermes and Milon(ius) in the graffiti
should be interpreted as the opus of the workshop “boss”
(officinator), with the periodic activity of the workers in his
charge. This task is linked to the manufacturing stage of the
Dressel 20 amphora globular bodies, with the belly upside
down during the drying process, before it was joined to the
rest of the parts. For the “Milonis / Sosumus” annotation we
15
Abad Casal 1975, Fig. 19.
16
RIB 2493.34.
17
Ponsich 1974, 99, no. 143.
18
Berni Millet 2008, 242.
BEZECZKY_Text.indd 128
EPIGRAPHY
could propose the following interpretative reading: “Pro-
duction of the pottery workshop of Milon(ius) worked by
Sosumus”, which would demonstrate the complexity of the
control system that was no doubt essential for the adminis-
tration of the figlina in the exercise of a concurrent, highly
industrialised activity. On the other hand, the readings “Cal-
listus f(ecit)”, “Hermes f(ecit)” and “Milon(ius) f(ecit)” on the
stamps only serve to reconfirm the skilled supervisory role of
the workshop foremen in managing the amphora manufac-
ture they had been charged with by the figlina owner.
From the epigraphic material excavated at Monte Tes-
taccio we know that the nominal relationship between the
graffiti and the stamps was neither constant over time, nor a
generalised practice in the manufacturing regions of Baetica.
Only a small number of coincidences has been recorded to
date and these are probably limited to the aforementioned
case of Villar de Brenes and then only in its later produc-
tion phases. Nevertheless, we can expect new peculiar cases
to arise, such as the one suggested for the VITALIS stamps
from the third century A.D. found at Monte Testaccio along
with a graffito with the same name, which would allow us to
consider the possibility of it being the same person.19
3 CALENDAR GRAFFITI
In the same position on the amphora belly, always upside
down and hidden, in cursive script, we find with relative
frequency the so-called calendar graffiti giving the date, the
month and sometimes even the year, normally preceded
and/or followed by simple names of persons declined in the
nominative or genitive. From our point of view these graffiti
19
García – Ozcáriz 2007.
18.03.19 14:40
 CHAPTER 10 – Calendar graffiti on Dressel 20 amphorae (Berni Millet)
are of the same nature as the nominal graffiti we have already
seen. We believe they would have been considered an exten-
sion of them, giving a more detailed record of the pottery’s
production of amphora bellies, in which it is important to
note the passage of time.20
In the epigraphist’s eyes, this cursive graffiti stands out for
its excess and flashiness, with one, two or three lines of writ-
ing, depending on the type of formula, and the date as the
linking element of the written message.21 Below we will see
which are currently the best known formulas and how they
link the related nominal elements.22
In the calendar graffiti formula polarised by two cogno-
mina, the first person’s name is always written in full and may
appear declined in the first or second case, whereas the other
name, when it is complete, is written in the nominative.
For example, “Melande[r] / iii k(alendas) octo[bres] / Gallo”
reminds us of the binominal graffiti with the two names
expressed in the nominative. The graffito “[Ma]rtin[us] /
[---] k(alendas) apriles / Meridi(anus)” adopts a similar for-
mula to the previous case, except in the detail of the abbrevi-
ated second name. On the other hand, the graffito “Asiatici
/ xiii k(alendas) sep(tembres) Luc( )” has the first cognomen
written in full in the genitive, while the other, Luc( ), is
abbreviated and would probably have been declined in the
nominative.
Rodríguez Almeida23 had already realised that there was
a systematic formulism in the composition of these records.
When he compiled and studied the copious Martinus series
of calendar graffiti found on Monte Testaccio in superfi-
cial contexts from the mid-second century A.D., he was
surprised to find that this person always occupied the first
line of the inscription, whereas the second name on the last
line was variable: “[M]artinus / iiii idus ianuarias / Gallo”,
20
The use of calendar graffiti in the figlinae accounting and
organisational system is well-documented in the area of industrialised
building material manufacture. In this respect, see the extremely
comprehensive catalogue of cases compiled by Scholz (2013, 351-
355). Also Matijašić’s (1986) epigraphic study of ante cocturam
graffiti on the lateres sesquipedales in the Zagreb Museum, where,
following the date in the Roman calendar we find the name of the
person responsible for keeping account of the bricks made by each
of the workers in their gang. In several of these inscriptions we find
the expressions “in hoc navali” and “in alio navali” in the colloquial
jargon of the potters referring to the drying areas where the recently
made clay objects would be placed. Matijašić also reminds us how
the bricks had to be left to dry for long periods in specially prepared
rooms before they could be fired.
21
Rodríguez Almeida 1993, 104-105.
22
For the bibliographical references cited below please see the general
table of calendar graffiti at the end of this paper.
23
Rodríguez Almeida 1991, 37.
BEZECZKY_Text.indd 129
129
“[Ma]rtin[us] / [---] k(alendas) apriles / Meridi(anus)”,
“Martinus / xii k(alendas) iunias / Tuscianus”. Moreover they
shared the same way of writing and composition, meaning
they must have been written by the same hand. Therefore
Martinus might have been the signatory because he is always
on the first line and the person below would have been a sub­
ordinate in his gang of workers. In the formation of the ductus,
the cursive writing technique with the stylus is perfect (Fig.
10.5) and shows that the author of the graffito was an expert
scribe, a person who was well qualified to hold a senior posi-
tion as the head or supervisor of a productive section in the
figlina.
Fig. 10.5 Calendar graffito from the first group: “Martinus / xiiii
k(alendas) sep(tembres) / Gallo” (Rodríguez Almeida 1991,
Fig. 6).
The Martinus case is in line with the procedure under-
taken by other signatories we will meet throughout this
paper. At this point we should ask ourselves how we can
interpret what lies behind the “cognomen + calendar date
+ cognomen” formula, taking into account the alternating
cases (nominative or genitive) in the first nominal element.
We believe this variation in grammatical declension does not
alter the message or change its meaning and is in fact two
alternative forms of expressing the same thing. According
to Rodríguez Almeida,24 when the name of the signatory
is expressed in the genitive case we can assume the implicit
presence of the noun opus; if it appears in the nominative
it would be the verb scripsit, and for the other various and
subordinate names it will always be fecit. In that case, we
would have the following interpretative readings: “Martinus
/ xii k(alendas) iunias / Tuscianus” = “Production controlled
by Martinus, deposited the twenty-first of May, worked by
Tuscianus”; “Asiatici / xiii k(alendas) sep(tembres) Luc( )” =
“Production of Asiaticus, deposited the twentieth of August,
24
Rodríguez Almeida 2000, 1287.
18.03.19 14:40
 130 EPIGRAPHY
worked by Luc( )”. However, we suggest that Martinus and
Asiaticus were not directly responsible for writing the graf-
fiti. They might instead be heads of two ceramist offices,
those responsible for directing the manufacture of the
amphorae. His subordinate would be a worker in charge of
the officinator. This type of inscription registered an activity
of the figlina, giving the name of the worker who had manu-
factured the group of globular bodies, put to dry in a section
of the workshop, with the date of delivery, plus the name of
the officinator for whom he had undertaken that task. The
true author of the graffito is anonymous, he did not write his
name, having to occupy a position of greater relevance in the
internal organization of the figlina. Among his responsibili-
ties would be to supervise the periodic work done in the cer-
amist offices when organizing the space in the drying areas.
There is a second group of two-line calendar graffiti in
which only one of the two persons alluded to is included, up
to now always found with their full name, expressed in the
genitive or the nominative and before or after the calendar
date. For all practical purposes this could give the impression
of a simplification of the previous formula, with the omis-
sion of the subordinate: “Menander / xiiii k(alendas) iulias”
BEZECZKY_Text.indd 130
Fig. 10.6
Three calendar graffiti from the
second group: (1) “iii k(alendas)
octobres / Fortunatus” (Geneva,
Paunier 1981) (2) “Menander
/ xiiii k(alendas) iulias” (Heer-
len, Photo from Radboud Uni-
versity) (3) “Menandri / v nonas
octobr(es)” (Tiel, Photo and sketch
from Radboud University).
(Fig. 10.6.2), “Menandri / v nonas octobr(es)” (Fig. 10.6.3),
“Martini / [---] k(alendas) maias”; or the omission of the
officinator (main character): “iii k(alendas) octobres / For-
tunatus” (Fig. 10.6.1). However, we cannot be completely
sure of this observation, as the principal protagonist is some-
times recorded after the date and the month, meaning that
it is possible that we are not dealing with the author of the
graffito.25 Look carefully at the two graffiti capitalized by
Menander (Fig. 10.6) found in the Dutch towns of Tiel and
Heerlen, with his name written irregularly in the nominative
or genitive.
It is interesting to corroborate the absence of calendar
graffiti from the first and second group in the first or third
century A.D. contexts. The Martinus graffiti collected by
Solin and Rodríguez Almeida come from the northern slope
of Monte Testaccio, where the surface finds give consular
dates from the mid-second century A.D. Our general table
of calendar graffiti in Section 7 also highlights this fact,
when we see that the dated material is limited to the second
century A.D. Moreover, as can be seen in the epigraphic
25
Rodríguez Almeida 1984, 265.
18.03.19 14:40
 CHAPTER 10 – Calendar graffiti on Dressel 20 amphorae (Berni Millet) 131

finds from the Monte Testaccio excavation campaigns, calen-

dar graffiti are only catalogued for the Antonine period and
their absence is notable in the huge amount of ante cocturam
graffiti recorded for the third century A.D.
Less frequent, but of great interest, are the graffiti from
the third group with the calendar date accompanied by the
consular date, plus the name of the officinator that always
ends the text in the second case. The best known examples
to date are composed on two lines of script and give us con-
sular dates from the mid-second century A.D., between 147
and 164 A.D. Several of them share the same palaeographic
features, allowing us to attribute to them the same signature
in different years of activity.
The most striking current example is that referring to one
Lucrio: “Tertullo et Sa[cerdote cos] / viiii k(alendas) iul(ias) Fig. 10.7 Calendar graffito by Lucrio with a consular date of 23 June
158 A.D.: “Tertullo et Sa[cerdote cos] / viiii k(alendas) iul(ias)
Lucrio[nis]” dated 23 June 158 A.D. (Valkenburg, Holland) Lucrio[nis]” (Valkenburg, Photo: Berni Millet).
(Fig. 10.7); this is clearly equivalent to the Dressel find26 rein-
terpreted by Rodríguez Almeida:27 “[Commodo et] Lat(e)
rano cos / [---] Lucrionis” from four years earlier (154 A.D.).
No less striking are the three Barbarus graffiti found in Swit-
zerland, with two mentions in Augst: “Praesente et Rufin[o
co]s / iii k(alendas) iun(ias) Barbari” from 30 May 153 A.D.,
“[---]+ [---] / [---]i k(alendas) i+[---]” on a small shard with
identical calligraphy;28 and a third from the Lenzburg vicus:
“[Orfito et Pri]sco cos / [---] k(alendas) apr(iles) Barbari” also
with four years between them (March or April 149 A.D.).
We can complete this modest list of known signatories with
the names Felix and Cato: “[Orf ]ito et Prisco cos/ [---] Feli-
Fig. 10.8 Calendar graffiti from the fourth group: (1) “vii id(us) apr(iles)”
cis” from 149 A.D. (Bearsden, Scotland),29 “Condiano cos / (London); (2) “kal(endis) novembri(bus)” (Hofheim); (3)
xviii k(alendas) octobr(es) Catonis?” from 15 September 151 “[---] nove[mbri(bus, -es)]” (Lincolnshire); (4) “<kalendis>
A.D. (Turin, Italy). We can also include three partial graf- novembri(bus)” (Digoin, Photo: Guillaume Maza).
fiti in which the name of the principal protagonist has been
lost, but which have the earliest and the latest dates currently This must have been a simplified variation of the preced-
known: “Largo e[t Messalino cos] / [---]++ +[---]” from 147 ing types with the same function in the productive process,
A.D. (Testaccio),30 “ii(duobus) Caesaribus c[os] / +[--- ---] despite the lack of workers’ names. Perhaps this atypical
nis” from 161 A.D. (Kempten, Germany),31 and “[Macrino formula can be attributed to an earlier period in the use of
et] Celso cos / +[--- ---]+ati” from 164 A.D. (Little Chester, calendar graffiti, taking into account the “vii id(us) apr(iles)”
England).32 inscription from London associated with a mid-first-century
Finally, we should distinguish a fourth group of cursive A.D. archaeological context.
graffiti with a simpler formula limited to the calendar date Another case with the date “kal(endis) novembri(bus)”
with no names and occupying a single line of text (Fig. 10.8). is listed in CIL, XIII as a find from near the Roman mili-
tary camp of Hofheim in Germania; it shows a ductus with a
clear correspondence to other graffiti from Digoin (France)
26
CIL, XV. 3614.
and Lincolnshire (England). The list is completed with
27
Rodríguez Almeida 1984, 256, Fig. 91.1.
28
Martin-Kilcher 1987, 71 (G2), Fig. 41. the unpublished graffito “xiii k(alendas) octobres” from the
29
RIB 2493.69; Burnham et al. 1997, 472 (e). “De Holdeurn” archaeological site in Nijmegen (Holland)
30
Rodríguez Almeida 1984, Fig. 92. (Fig. 10.9).
31
Schimmer 2009, Pl. 18.629.
32
RIB 2493.8.

BEZECZKY_Text.indd 131 18.03.19 14:40

 132
Fig. 10.9 An unpublished graffito from the fourth group: “xiii k(alendas)
octobres” giving the date of 19 September (“De Holdeurn”,
Nijmegen, Photo: Radboud University).
Equally as unusual as the calendar inscriptions from the
fourth group is the following find from Monte Testaccio:
“Muciano et Fabiano cos”, with the single consular date from
the year 201 A.D.33 To date this type of graffito with consular
dating without the date of the month is unique. The absolute
dating places it in a much later period than the graffiti of the
third group belonging to the Antonine period. Thus, this
particular epigraphic formula appears to be limited to a dif-
ferent chronological horizon, perhaps later.
By looking at the different formulas found to date, we
have seen in detail the functioning of the most important
calendar graffiti from the four general groups. This grouping
is probably not yet definitive and in the future we may well
be able to add new combinations based on other procedures
adopted by the authors of these records. Now we have to ask
ourselves what the true purpose of this type of annotation
would have been in the productive economy of the figlinae
that produced the Dressel 20 amphorae.
The particular position of these “hidden” graffiti, with the
text inverted on the base of the amphora, is the main argu-
ment used to suggest they were linked to the manufacturing
and drying phase of the globular bellies. It is also obvious
that when they were placed in the drying room, a senior,
well-educated worker recorded on some of them the date of
reception of the batch of bellies manufactured by a subordi-
nate in the pottery.
From our point of view, the use of graffiti with the calen-
dar date is evidence of the need to compute the amount of
time the amphora bellies needed to be left in the specially
prepared and protected drying area.34 It could take sev-
33
Testaccio 2010b, 301, no. 517.
34
Berni Millet 2008, 35.
BEZECZKY_Text.indd 132
EPIGRAPHY
eral weeks for the clay of the bellies to dry, as the spherical
shape and thickness of the walls required a slow and care-
fully monitored process. Only in this way was it possible
to obtain hard, compact bellies that could be moved on to
the next step in the production line. This hypothesis is sup-
ported by an exceptional find preserved in the Museum of
Aoste (Isère), an anecdotic three-line graffito written in rus-
tic capitals on a shard of Dressel 20 amphora in which we
can read explicitly:35 “Gratus / XX dies / argilam”. The find
from Aoste, a vicus in the city of Vienne in the Rhône-Alpes
region, records the name of a worker – Gratus –, and has
been read as “Gratus, XX dies (ad) argilam”. The explanation
given by scholars is that it is an account of the work done
by an unknown potter working for Gratus. The position of
the inscription on the amphora and what is known from
the organization of the production of Dressel 20 amphorae,
especially the drying, suggests that this number could have
been a time reminder: “Gratus, leave it dry 20 days”. In the
same way, the date of deposit recorded by the “calendarial”
graffiti would have helped calculate the drying time needed,
depending on the climate and the time of the year. However,
this is only a hypothesis as we lack any other similar example
from the epigraphy of the Baetican province. Another pos-
sible reading could be: “Gratus (laboravit) XX dies (ad) arg-
ilam”, suggesting that Gratus the slave was doing other tasks
in the kiln, such as collecting the clay. Following this third
reading, the potters from Baetica linked to some kiln would
had their own assistants to do complementary tasks like the
preparation of the clay itself.
4 THE CALENDAR GRAFFITO FROM THE
CASTRUM OF BRIJUNI
The ante cocturam calendar graffito from the Castrum of Bri-
juni is preserved in full next to the basal spike on the large
globular belly of a second century A.D. Dressel 20 amphora
(Fig. 10.10). It was probably already fragmented when it was
found, but it has been possible to restore it in full with large
pieces of the wall. The upper part of the Dressel 20 amphora
(belly, neck, rim and handles) is missing, as it had already
been cut off in Antiquity to reuse the large spherical body for
another purpose, in this case as a fixed, immovable object in
the style of a small dolium.
The reuse of Dressel 20 amphora globular bodies was a
fairly habitual practice in the markets to which they were
exported. Its peculiar spherical form, with thick walls that
35
Rémy – Jospin 1998, no. 1.
18.03.19 14:40
 CHAPTER 10 – Calendar graffiti on Dressel 20 amphorae (Berni Millet) 133

Fig. 10.10 The large globular body of

the Dressel 20 amphora from
Brijuni with an ante cocturam
calendar graffito on its base
(Photo: Bezeczky; Sketch:
González Cesteros).

Fig. 10.11 (1) The disposable upper part of a Dressel 20 amphora found in Nijmegen (Bureau Archeologie in Monu-
menten Gemeente Nijmegen; Photo: Berni Millet); (2-5) Cut bellies with post cocturam graffiti from Augst
and Kaiseraugst (Martin-Kilcher 1987).

stood up well to pressure and knocks, meant that this belly
was much appreciated in the “recycling culture” that was
common in Roman society. In the case of the Brijuni find,
it had been used like a small dolium without a flat bottom,
meaning that it must have been partially buried in a hole or
placed on some kind of stand to maintain it in a stable verti-
cal position.
There is no shortage of examples that illustrate the reuse
of these amphorae in any of the markets that consumed olive
oil from Baetica. The number of published sketches makes
the finds from the Roman settlements of Augst (Augusta
Raurica) and Kaiseraugst near Basel (Switzerland) – studied
by Stefanie Martin-Kilcher36 – a good guide. In them we can
36
Martin-Kilcher 1987.
see two alternative ways of cutting the vessel, depending on
the diameter of the mouth required (Fig. 10.11). A narrow-
mouthed vessel was obtained by cutting the amphora at the
height of the beginning of the neck and the bottom of the
handles (Fig. 10.11.4). To obtain a wide mouth the upper
body of the amphora was removed several centimetres below
the connection of the two handles (Fig. 10.11.5). The choice
of where to make the cut may have depended on the type
of product to be stored in it. A wide-mouthed container
would have been more suitable for filling with or emptying
out solid substances, while a narrow mouth would have been
better for liquids. With the Dressel 20 amphora from Bri-
juni the narrow-mouthed cut was chosen, giving an aperture
of 22.6 cm in diameter. This provided a small dolium with a
diameter of 52 cm, a height of 58.5 cm, 2.5 cm-thick walls

BEZECZKY_Text.indd 133 18.03.19 14:40

 134 EPIGRAPHY

and a capacity of approximately 8 modii (70.32 litres), the

equivalent of 89.29 % of the total standard capacity (78.75
litres) of a mid-second century A.D. Dressel 20 amphora.
Another important detail pertaining to the recycled items
from Augst and Kaiseraugst is the high incidence of post coc-
turam graffiti near the line of the cut, along the edge of the
new mouth of the vessel or what remains of the beginning of
the lower part of the amphora handle (Fig. 10.11.2-3). They
are almost always numeric and nominal indications that con-
tinue to remind us of the modus operandi with large dolia.
These post cocturam inscriptions refer to the new capacity of
the cut down globular body (given in modii and sextarii) and
the final recipient of the amphora turned into a dolium. On
the Brijuni Dressel 20 amphora there are no signs of post coc-
turam writing that would allow us to evaluate this.
Returning to the Brijuni find, the loss of the rim and han-
dles makes it much more difficult to obtain a precise relative
dating for its manufacture in Baetica, taking into account the
two most characteristic morphological features studied for
the evolutionary model of the Dressel 20 amphora.37 Nev-
ertheless, we can narrow it down to a point in the second
Fig. 10.12 Calendar graffito on the base of a Dressel 20 amphora from
century A.D., bearing in mind the full profile of the globular
Brijuni (Photo: Bezeczky).
body and the modest size of the spike. We know that first cen-
tury A.D. Dressel 20 amphorae developed a spherical body
with a more regular profile together with a more prominent
conical spike. This is not the case with those produced in
the third century A.D., which are characterised by a dispro-
portionate design, with the foot reduced to the minimum
expression and an irregular spherical body with its maximum
horizontal diameter nearer the amphora base. The Brijuni
amphora’s body (Fig. 10.10) has the distinctive morphologi-
cal features characteristic of those manufactured in Baetica
during the Antonine period. The spike is conical but small, Fig. 10.13 Brijuni calendar graffito (Sketch: Berni Millet).

practically reduced to a simple button (Fig. 10.12). The max-

imum diameter of the body is in the centre of the circumfer-
ence, although its complete profile does not form a perfect
sphere, as there is a slight carination at the join between the
cylindrical body and the upper part of the amphora. In sec-
ond century A.D. productions it is possible to see clearly the
flexion at an obtuse angle created at the moment the two
parts of the Dressel 20 amphora were assembled. This appre-
ciation is also corroborated by the typology of the calendar
graffito on the Brijuni amphora which, as we will see, can
also be confined to this second century historical context.
The Brijuni ante cocturam calendar graffito is inscribed
next to the foot, on the bottom of the Dressel 20 amphora
37
Berni Millet 1998, 33-53; Berni Millet 2008, 58.
belly (Fig. 10.12). The inscription has been preserved almost
intact, with just a small diagonal fracture that does not
impede its full reading (Fig. 10.13):
Asiatici
sec( ) x k(alendas) mar(tias) Alb( )
The text of the calendar graffito with its protagonists is
written on two lines. Below, to the right, the figure II has
been added obliquely, perhaps to indicate that it was part of
the second batch or second row of bellies finished by Alb( )
from the officina of Asiaticus that day. This last detail is

BEZECZKY_Text.indd 134 18.03.19 14:40

 CHAPTER 10 – Calendar graffiti on Dressel 20 amphorae (Berni Millet) 135

important, because it means the cursive graffito refers to a 5 ASIATICUS: ANOTHER PARADIGMATIC
set of bellies and not just this one. CASE
The content of the cursive inscription is structured
Asiaticus’ epigraphic repertory is a new paradigmatic exam-
around the calendar date as a central element linking the
ple similar to those of Martinus and other graffiti from
whole written message. The Roman calendar date given is 20
Monte Testaccio, given the high number of cases recorded
February: “(ante diem) decimum k(alendas) mar(tias)” (“ten
for the approximately fifty examples of calendar graffiti cur-
days before the calends of March”). The text is opened and
rently known (see Section 7). It was a pleasant surprise to
closed by given names. The first, Asiaticus, is the officinator
be able to assemble seven graffiti capitalized by this person,
and is written in the genitive case and the abbreviated cog-
counting the Istrian example. We were also surprised by the
nomen Alb( ) would have been the subordinate craftsman
wide geographical spread of the finds, with four examples in
(figulus) who made the bellies in Row II; completed in the
Britannia, three in Gallia and one in Germania (Fig. 10.14).
nominative case it would have read Albinus or Albanus.
Knowing the signature of the author of the graffito, the spe-
We also have to note an anomaly in the formula of the graf-
cial features of the writing and the content formula is key to
fito, with the intrusion of the abbreviated word sec( ) at the
resolving the frequent challenges we come up against when
beginning of the second line, preceding the date; this word
interpreting the texts, such as those detected in the graffiti of
is in an unusual position and we are unsure of its meaning.
Asiaticus in the publications cited below.
In the opinion of Ivan Di Stefano, to whom we are grate-
Asiaticus always used the same epigraphic formula in his
ful for his valuable assistance in this matter, the abbreviated
calendar graffiti. The inscription occupies two lines of writ-
word preceding the date of the month could be understood
ing, the first with his full name in the genitive and the sec-
as “sec(undum) (opus)”, to indicate “due cicli lavorativi cro-
ond with the date of the month preceding the abbreviated
nologicamente contemporanei”. In that case this abbreviated
cognomen of a subordinate. Only in the Brijuni and Alten-
word would be a transliterated repetition of the number II
stadt graffiti do we find that strange abbreviated word sec( )
that accompanies the cursive graffito. We will find a further
preceding the date of the month; in all the others it is absent.
instance of the abbreviated word sec( ) in another graffito
The months were abbreviated with the first three letters:
capitalized by Asiaticus and found in the German locality of
ian(uarias), mar(tias), Sep(tembres). From the full dates we
Altenstadt (cf. Section 5.1).
know they worked, at the very least, between the months of
The Brijuni ante cocturam graffito was written with a sty-
lus. The use of this instrument on damp clay conditions the
palaeographic lettering of the cursive script, as it obliges the
hand to make undulating movements that vastly favour the
progression of the vertical, inclined and curved strokes.38
We can make some interesting calligraphic observations
to determine those features that characterised the author’s
writing. First of all, note the almost rectilinear movement of
the letter ‘s’ in the words Asiatici and Sec( ), with an appar-
ently long, oblique ‘i’. The archaic form of the letter ‘e’ in the
abbreviated word sec( ) on the second line is inscribed with a
double vertical curved stroke, a common solution in cursive
script written with a stylus. The letter ‘a’ in the abbreviated
word mar(tias) is also written with the same angular stroke
as the letter ‘m’. The inscription’s written message is in the
typical formulaic structure of the first group of calendar graf-
fiti from the mid-second century A.D.

38
Rodríguez Almeida 1991, 38-40. Fig. 10.14 Geographical distribution of the Asiaticus graffiti.

BEZECZKY_Text.indd 135 18.03.19 14:40

 136 EPIGRAPHY

January and December and we know three of the workers

under him from the abbreviated names Alb( ), Luc( ) and
Amat( ).
Asiaticus’ graffiti can be grouped into two series from
the clearly differentiated style in which he wrote his name,
probably because different hands or individuals made these
records in the figlina. The formation of the letters is more
regular, spaced and recognisable in the examples from Bri-
juni, Altenstadt, Lincoln and Wall. In the Saint-Thibéry,
Reims and Brough on Noe graffiti these qualities are altered
and the writing is irregular, more careless and compact, mak-
ing it quite difficult to understand.
The formula of Asiaticus’ calendar graffiti dates them to
the Antonine period, based above all on the inscriptions
from other similar persons that are well dated at Monte
Testaccio in surface levels from the mid-second century
A.D. The profile of the Brijuni amphora globular body can
be typologically limited to that period in the second century
A.D. The rest of the inscriptions lack precise relative dating,
but there are indications that point to this horizon, consider-
ing the archaeological context of some of the finds.
5.1 ALTENSTADT (HESSEN, GERMANY)
The first example comes from the German town of Alten-
stadt, near which a Roman auxiliary camp was built during
Domitian’s military campaigns in that country. It remained
active until the German-Rhaetian limes were abandoned in
the mid-third century A.D.
The graffito is preserved fragmented on a piece of amphora
wall. The inscription was initially published with the read-
ing “ASI.../SIICIIIIF...”.39 It has recently been re-published
in a monographic study of amphorae from the Roman
archaeological sites in the area of influence of Mogontia-
cum (Mainz),40 the capital of the new province of Germania
Superior established by Domitian. In this latest study, the
Dressel 20 form of the ante cocturam graffito is identified by
a new sketch, although there is no transcription or proposal
for its reading.
The lost text can be largely restored thanks to the paral-
lel found in full in Brijuni: “Asi[atici] / sec( ) iiii k(alendas)
i[an(uarias), -un(ias), -ul(ias)] [---]”. There can be no doubt
that it was signed by our Asiaticus, as the composition, for-
mula and calligraphy are the same. Moreover, once again
we come across that strange, unknown abbreviation sec( )
39
Schönberger – Simon 1983, 161 no. 55, Pl. 68.D55.
40
Ehmig 2007, Pl. 1.169.
Fig. 10.15 The Altenstadt graffito (Schönberger – Simon 1983,
Tafel 68. D55).
preceding the date of the month and beginning the text on
the second line. The first letter ‘i’ of the month is preserved,
meaning that the date must be 29 December, 29 May or 28
June. We are also missing the name of the potter, which was
probably abbreviated with the first three letters at the end of
the second line.
5.2 LINCOLN (LINCOLNSHIRE, ENGLAND)
The second example is from the Roman town of Lindum
(Colonia Domitiana Lindensium) in northern Britannia,
today the city of Lincoln in the county of Lincolnshire. The
inscription was discovered in 1959 in a disturbed archaeo-
logical level inside the southern tower of the Eastern Gate in
the upper part of the Roman colonia. The pottery from the
upper fill levels is dated to a time no earlier than the period
of Hadrian, with a strong presence of material manufactured
during the Antonine dynasty.41
The sketch and the photograph of the cursive graffito
from Lincoln were published in RIB 2493.6 (Plate VI) with
the transcription “A?ITALICI / XIII K SIIP LVC[---]”. The
text is interpreted as follows: “A(...) Italici / (ante diem) xiii
k(alendas) Sep(tembres) Luc[... et ... co(n)s(ulibus)]”; or A( )
Italicus the manufacturer of the amphora, made thirteen
days before the calendas of September (20 August), during
an undetermined consulate in which only the initial letters
of the name of the first magistrate are preserved.
The photograph of the graffito is sufficiently clear and
sharp for us to confirm that the text is complete. Likewise,
41
Wright 1960 221, 241 no. 35.

BEZECZKY_Text.indd 136 18.03.19 14:40

 CHAPTER 10 – Calendar graffiti on Dressel 20 amphorae (Berni Millet)
Fig. 10.16 The Lincoln graffito (RIB 2493.6).
it is of great help in corroborating some of the inaccuracies
in the irregular form of reproducing the letters ‘i’ and ‘t’ in
the name Asiaticus in the sketch. With our knowledge of
the Brijuni graffito, this one is easy to read: “Asiatici / xiii
k(alendas) sep(tembres) Luc( )”. Once again it is written on
two lines, with the same type of lettering and has a practi-
cally identical formula to those already seen. There are only
two substantial differences in the content. The first is a new
subordinate called Luc( ), written in an abbreviated form
with the same length of letters; the second is the omission
of the anomalous word sec( ) preceding the Roman calendar
date.
5.3 WALL (STAFFORDSHIRE, ENGLAND)
The third example was found in 1925 in the ancient Roman
settlement Letocetum, near the town of Lichfield in the
county of Staffordshire. The archaeological site is in the par-
ish of Wall. In Antiquity it was an important military outpost
and travellers’ rest stop that had been founded at a strategic
crossroads in the early years of the invasion of Britannia. The
settlement reached its high point during the second century
and flourished during a large part of the third century A.D.
The amphora graffito was first published with the tran-
scription “NI·ATICI / XVII K I...”.42 It has subsequently
been republished as “ASI·ATICI / XVII K I[---]”43 with
the correct reading: “Asiatici / xvii k(alendas) i[an(uarias),
-un(ias), -ul(ias)] [---]” (16 December, 16 May, 15 June).
Note in the sketch the almost rectilinear stroke that is so
characteristic of Asiaticus’ letter ‘s’, in this case touching
the ‘A’, which gives the visually false impression of an ‘N’.
The interpunctuation mark is accidental or an impurity in
the sketch. This is yet another graffito capitalized by our
42
Taylor – Collingwood 1925, 250 no. 22.
43
RIB 2493.5.
BEZECZKY_Text.indd 137
137
Fig. 10.17 The Wall graffito (RIB 2493.5).
Asiaticus, once again on two lines with the same formula
and composition. Obviously, we are missing the abbreviated
name of the subordinate after the date of the month and,
for the second time, there is no trace of the anomalous word
sec( ) found in the Brijuni and Altenstadt graffiti.
5.4 BROUGH ON NOE (DERBYSHIRE,
ENGLAND)
The following ante cocturam graffito is on the base of a glob-
ular amphora excavated in 1962 in a ditch alongside the
Roman road of Batham Gate, just outside the southeastern
gate of the military encampment of Navio.44 The Roman
settlement is to the east of the hamlet of Brough on Noe
in the county of Derbyshire. The Roman auxiliary fort was
built in the mid-first century A.D. and rebuilt in the time
of Antoninus Pius. There is evidence of destruction around
196 A.D. and the modification of its internal structures in
the time of Severus.
The inscription was first published with the rather strange
reading of “ILITRIIB[---] / LIII F?RVI?[---] / [---]I[---]”.
Later, the sketch of the graffito was published with a differ-
ent transcription that was closer to the truth:45 “[---]TRII+[-
--] / VIII K AV[---] / [---]+[---]”; reading the date as 25 July.
The sketch of the graffito is quite simple and, in general,
a poor reproduction of the inscription details. Even so, it is
possible to read the controversial name on the first line by
counting the strokes of the possible letters: “Asiatici / viii
k(alendas) au[g(ustas) ---]”. The calligraphy is not as even as
that we have seen up until now. The shafts of the letters tend
to lean towards the left due to the support point of the hand
being forced. Once again the first two letters –‘A’ and ‘s’–
give the false impression of an ‘N’. We can assume that the
44
Wright – Wilson 1963, 166 no. 50.
45
RIB 2493.46.
18.03.19 14:40
 138
Fig. 10.18 The Brough on Noe graffito (RIB 2493.46).
missing piece of the shaft in the middle of the third line did
not exist as such, in view of the two-line formula of Asiati-
cus’ graffiti.
5.5 SAINT-THIBÉRY (HÉRAULT, FRANCE)
The fifth example comes to us from the French town of
Saint-Thibéry, in the Languedoc-Roussillon region, near
the Roman bridge that allowed the Via Domitia to cross
the department of Hérault. The place of the find is given as
“Nadailhan”, which must be the agricultural estate of Nad-
alhan, the site of the Benedictine abbey of Saint-Thibéry, an
area with Roman remains associated with a Gallo-Roman
villa.
Fig. 10.19 The Saint-Thibéry graffito (Rouquette 1980, no. 28).
The ante cocturam graffito is broken seven centimetres
from the point of the Dressel 20 amphora foot.46 Only the
first line is legible: “Asiatici / [---]++[---]++”, with the name
Asiaticus accurately interpreted in the publication, despite
the difficulty in the composition of the letters. The calli-
graphic qualities and characteristics of this inscription are
similar to those of the following example from Reims, in
which we describe the principal functionalities.
46
Rouquette 1980, 156 no. 28.
BEZECZKY_Text.indd 138
EPIGRAPHY
Fig. 10.20 The Reims graffito (photograph courtesy of Xavier Deru).
5.6 REIMS (MARNE, FRANCE)
The next example on the list of Asiaticus’ graffiti is an unpub-
lished find from the French city of Reims, provided to us by
Xavier Deru for its reading and publication. It was found in
2003 during excavations directed by Ph. Rollet (Inrap) in the
Rue de l’Equerre. As there is currently no definitive study of
the excavation, we lack a dating for the archaeological con-
text (3526) in which the find was made.
The inscription is of high quality and preserved intact on
the base of the Dressel 20 amphora with the surface layer
of clay in the process of exfoliation. We read: “Asiatici /
vii id(us) ian(uarias) Amat( )”, with the date of 7 January.
Note the rectilinear scored stroke added on the right hand
side half way up between the two lines. We interpret this as
a number with the same accounting function as the ordinal
numeral II documented next to the Brijuni cursive graffito.
The order and lettering of the words is always the same: the
signatory on the first line, the name of the month indicated
by three letters, the subordinate Amat( ) abbreviated and
ending the inscription on the second line.
The cursive lettering in the name Asiatici shows an aggres-
sive style with evident similarities to the Saint-Thibéry graf-
fito in the form of the ductus. The strokes of the letters tend
to lean towards the left, as we can clearly see from the ‘aes’
in both of them. Another particularly appreciable feature is
the rectilinear ‘s’ that cuts through the second shaft of the ‘A’,
resulting in the aforementioned false ‘N’. Note also how the
last ‘I’ is joined to the letter ‘c’, as if it were a ‘g’.
18.03.19 14:40
 CHAPTER 10 – Calendar graffiti on Dressel 20 amphorae (Berni Millet) 139

Fig. 10.21 Left: The Béziers graffito (cour-

tesy of Raphaël Macario);
Right: The Vindolanda graffito
(courtesy of Andrew Birley and
Barbara Birley).

5.7-8 ADDENDUM: BÉZIERS (HÉRAULT, The Asiaticus graffiti containing a readable calendrical
FRANCE) AND VINDOLANDA date are spread throughout the year, on 7 January, 20 Febru-
(BARDON MILL, ENGLAND) ary, 25 July, 20 August, and now 28 December. This dataset
shows that we are facing an important manager of a pro-
With our study formally completed, we once again open
duction centre of pottery, where he worked in a permanent
this section in order to give the news of two new Asiaticus
office specialized in the production of amphorae, in no way
graffiti, recently recovered in two archaeological excavations
an occasional or temporary activity.
carried out in England and France. With these two addi-
tions, we account a total of eight cases, revealing yet again
an astonishing geographical spread of this kind of epigraphic 6 GENERAL REFLECTIONS
document under a given individual signature.
By way of conclusion, we end this overview of calendar graf-
I have personally been informed of the French discovery
fiti with some reflections that may be of general interest for
(Fig. 10.21, left) by the archaeologist Raphaël Macario, and
future research into the subject.
I am very grateful for the privilege of having been allowed
The first, and that which has most attracted our attention,
to publish this information. The piece was collected in 2014
is the high rate of repetition among the principal protago-
during construction work carried out in the centre of the
nists of calendar graffiti, whether seen through the exam-
town of Béziers. This small pottery shard partially preserves
ples found in a specific geographical area – as is the case of
the original text of a graffito consisting of two lines of writ-
Martinus at Monte Testaccio in Rome – or the widespread
ing and the typical epigraphic formula: Asiatic[i] / [---]i ++
continental distribution of Asiaticus’ graffiti. We can recall
[----]. The Asiaticus name reappears with the first two letters
the five Martinus graffiti from Monte Testaccio with the
connected and the rectilinear S, thus giving the false appar-
same formula and different partners: Gallo (10 January and
ent form of a letter N, as we saw in previous examples. Below
19 August) and Tuscianus (21 and 27 May), both repeated
this some traces of the lost mutilated calendrical date are dis-
twice with different dates, and Meridi(anus) (16-31 March)
cernible, which might precede the subordinate character of
with only one example. Likewise, the seven graffiti capital-
the end of the second line.
ized by Asiaticus found in Croatia, France, Germany and
The English graffito (Fig. 10.21, right) comes from the
England with another list of subordinate workers: Alb( ) (20
Vindolanda archaeological campaign of 2015 and has come
February), Amat( ) (7 January) and Luc( ) (20 August). The
down to us through the kindness of the excavation directors
sum of these two series represents 30 % of the total number
Andrew Birley and Barbara Birley. This artefact (SF19409 –
of examples attributable to the first two of the groups of cal-
context V15-3B), a body sherd of amphora, was located
endar graffiti known today, the proliferation of which never
below a Severan forts drain foundation, lying on the crushed
fails to surprise us. This curious set of coincidences seems to
sandstone and clay floor of a Antonine workshop dated ca.
have no end, as it can also be extrapolated to other signato-
180-208 A.D. This is, again, another inscription partially
ries who are not so well documented by epigraphic finds.
preserved with the initial part of the text in its only two lines,
and can be clearly read: “Asia[tici] / v k(alendas) ian[uarias
----]” (28 December).

BEZECZKY_Text.indd 139 18.03.19 14:40

 140
We recently reported on the first calendar graffito exca-
vated in situ in a production area in Baetica,47 at the pottery
of Adelfa near the Roman towns of Carmo (Carmona) and
Canama (Alcolea del Río). The cursive inscription is writ-
ten on three lines of script on a small shard of a Dressel 20
amphora base. We read it as (Fig. 10.22.1): “Melande[r] /
iii k(alendas) octo[bres] / Gallo” from 29 September. This
is an exceptional find as it allows us to link the epigraphic
data to its place of origin, comparing its nominal elements
to the stamps from Adelfa, although we have yet to obtain
any positive results. Melander’s graffiti use the same formula
and composition we see in Martinus’ graffiti. We are also
surprised to find the name of Gallo again on the last line.
We need to ask ourselves if all these coincidences are pure
chance or if they are linked to a specific productive model.
An interesting suggestion would be that both officinatores
worked in the Adelfa figlina, coming to share a common or
interchangeable group of workers.
This appears to be a reasonable hypothesis if we take into
account the two Melander examples from Rome with a
chronological horizon close to that of Martinus. The former
is published in CIL, XV. 3609 (Fig. 10.22.2) as a find from
Fig. 10.22 Melander graffiti: (1) Adelfa (Moros – Berni Millet
2011, Fig. 11); (2) Testaccio (CIL, XV. 3609); (3) Testaccio
(Rodríguez Almeida 1991, Fig. 5A).
47
Moros – Berni Millet 2011, 41.
BEZECZKY_Text.indd 140
EPIGRAPHY
Monte Testaccio; it lacks a topographic coordinate, the text
is fragmented and the signature is attributed by Dressel to
[Ale]xander. Rodríguez Almeida48 believes it to be a graf-
fito written by a Menander, Alexander, etc., who wrote and
composed using the same system as Martinus. Now, having
seen the complete graffito from Adelfa and having observed
its calligraphic traits, we can recognise and correct the read-
ing of [Me]lander in Dressel’s find to: “[Me]lander / [--- iu]
nias / [---]ilis”. We have another graffito written in the same
style, although it is much more defaced and only preserves
the last letters of the name [Mela]nder on the first line
(Fig. 10.22.3). Rodríguez Almeida excavated it at his topo-
graphic coordinate C-7, where the surface epigraphic finds
are dated to 149 A.D., the same chronological context as the
Martinus surface finds at Monte Testaccio (see Section 7).
How can we explain this strange coincidence of repetitive
cases in a relatively short space of time? The only plausible
explanation in the face of such fortuitousness is to consider
an epigraphic phenomenon that was restricted exclusively to
specific figlinae during certain phases of the industrial activ-
ity. This is precisely what we saw with the binominal graf-
fiti from Villar de Brenes, which are all linked to the modus
operandi of a group of people who were working during the
final productive phase of the figlina in the periods of Marcus
Aurelius and Commodus.
As a second general reflection we consider another eco-
nomic question that has been little analysed to date. This is
the industrial organisation of the figlinae in relation to the
need and the capacity to produce surplus amphorae.
The calendar graffiti we know of give us dates that cover
the whole year (Fig. 10.23). Given that these inscriptions
were written when the Dressel 20 amphorae were being
manufactured, it is obvious that this activity was continuous
and uninterrupted.49 There are no idle months in our gen-
eral chart of calendar graffiti (see Section 7). We have dates
that coincide with the seasonal operations of harvesting the
olives and filling the amphorae with olive oil, which would
have impacted on the availability of the work force. The busi-
est months appear to be from May to October. Summer was
the busiest time for commercial activity, to compensate for
the number of amphorae in stock and to satisfy the strong
demand for containers. Autumn, a quiet period, was taken
advantage of to restock the warehouses with amphorae ready
for the logistic needs of the coming year.
48
Rodríguez Almeida 1991, 37.
49
Rodríguez Almeida 1984, 256.
18.03.19 14:40
 CHAPTER 10 – Calendar graffiti on Dressel 20 amphorae (Berni Millet)
Fig. 10.23 Definite dates for all the months of the year.
The reason for the incessant movement of containers
is obvious when you are aware of the huge annual produc-
tion of Baetican olive oil, which was of considerable impor-
tance in sustaining the economic and political system of the
Roman Empire.50 It was an unceasing business that involved
the movement of hundreds of thousands of amphorae every
year. It was promoted by public and private interests and had
large markets of civil and military consumers that were abso-
lutely subsidiary to and dependent on this economy. Bae-
tican olive oil production would have been of such a mag-
nitude that it would have needed a whole army of workers
devoted exclusively to meeting the huge demand for ampho-
rae that came with the annual diffusio olearia.51
Having reached this point, we have to consider the follow-
ing question: what was the economic criterion used in the
Baetican figlinae to calculate the expected annual demand
for amphorae? Taking into account the importance and
magnitude of such a stable, thriving business that was pro-
tected by the Roman state, we believe the production and
export of Baetican olive oil would have been independent of
the clients’ demand for containers. In other words, ampho-
rae would not have been manufactured according to the
needs of each client; instead forecasts would have been made
on the basis of the production and storage capacity of each
figlina. It would have been up to the managers of those fac-
tories to satisfy the needs of their customers by keeping their
warehouses full of finished amphorae, compensating for the
differences between sales and production by manufacturing
and replacing stock. This modus operandi of manufacturing
more containers than required by the production of olive oil
would have ensured that demand could be met at all times.
50
Remesal Rodríguez 1998.
51
Berni Millet – Gorostidi 2013, 186.
BEZECZKY_Text.indd 141
141
Not long ago we defended the hypothesis of the stockage
of containers in Baetica to explain the time lag between the
date the amphora was manufactured and the date it was used
seen on some of the finds from Monte Testaccio.52 Logically,
an amphora made in September would have remained in
the figlina warehouse until the following season, at least six
months if it was sold in the month of March. In fact, the few
cases of graffiti from Monte Testaccio we know of with con-
sular dates show a much longer time lag.
Rodríguez Almeida noticed this strange phenomenon
in the epigraphic finds from his 1990 excavation campaign.
That year he found two calendar graffiti with consular dates
of 151 (“Condiano et Maximo cos”) in a level dumped with
tituli δ from the years 153 (“Praesente el Rufino cos”) and 154
A.D. (“Commodo et Laterano cos”). At first he thought this
difference of a couple of years between the date the container
was manufactured and the date the olive oil was transported
were isolated cases of the use of old stock.53 He subsequently
proposed the concept of “surplus” stock in the amphora
stores as an explanation.54
The 1999 excavation campaign came up with another graf-
fito with a consular date of 201 A.D. (“Muciano et Fabiano
cos”) in a dump level with tituli δ from 207-208 A.D. for which
the time lapse is much greater – six years. Remesal55 explains
this time lag as chance, rather than stockage in its modern
meaning. He attributes it to a surplus of amphorae that would
have been small in comparison to the amount manufactured
annually. He explains that the number of amphorae manufac-
tured would have been slightly higher than that needed for the
distribution of olive oil in order to guarantee a supply for cus-
tomers in the case of unexpected demand caused by breakages
or perhaps an unusually high olive oil production. He does
not, however, explain the time lapse of six years. If the surplus
stock was small, how can we explain the fact that this amphora
spent such a long time in the figlina warehouse before it was
used to transport olive oil to Rome?
We believe a possible historical explanation can be found
by analysing the Severian economic policy in Baetica.
The authors of the Historia Augusta wrote about the large
amounts of food distributed by Septimius Severus during his
mandate, including abundant free rations of olive oil from
Baetica and Tripolitania (SHA Sev. 8.5; 18.3; 23.2). The
epigraphy on Dressel 20 amphorae corroborates this passage
52
Berni Millet 2008, 37.
53
Rodríguez Almeida 1993, 105 and note 38.
54
Rodríguez Almeida 2000, 1289.
55
Testaccio 2010b, 255-257.
18.03.19 14:40
 142 EPIGRAPHY
with Imperial family stamps and tituli picti. They are proof
of state intervention in the economy and Baetican olive oil
production and sale following Severus’ confiscations in Bae-
tica.56 On the other hand, the highest proportions of ante
cocturam stamps and graffiti found at Monte Testaccio are
on third century A.D. amphorae, in our opinion as a con-
sequence of a much more intensive mass production. This
allows for the possibility that the six-year time lapse seen in
the Monte Testaccio graffito was due to an increased need to
store surplus amphorae during the third century A.D.
The three graffiti with consular dates from Monte Testac-
cio may appear to be sparse evidence on which to definitively
resolve this controversial debate, but we have other options
to continue investigating this subject by comparing chro-
nologies.
The Imperial stamps of Septimius Severus’ family provide
three successive relative datings in the “augustorum nostro-
rum trium” (209-211 A.D.), “augustorum nostrorum duo-
rum” (211-212 A.D.) and “Augusti nostri” (212-217 A.D.)
series. Some years ago we compared the relative dating of
these Severian stamps to the contextual dates obtained from
the tituli picti at Monte Testaccio.57 We were astonished to
find time lags of up to seven years in some examples, a phe-
nomenon that was present in the epigraphy of both Baetican
and Tripolitanian olive oil amphorae.
The most significant case is that of a Dressel 20 amphora
excavated in the 1992 campaign; on its handle was the stamp
AVG[[GG]]N[[NNN]] from the Caracalla series (212-217
A.D.) physically associated with a titulus δ from the year 224
A.D.58 (Fig. 10.24). That same excavation campaign pro-
vided another example of the AVG[[G]]GNN[[[N]] Geta
and Caracalla series (211-212 A.D.) in a stratum dated by
the tituli picti to 220-221 A.D.59 The time lapse is consider-
able, as both amphorae were shipped many years after they
had been manufactured.
In the light of this evidence, Remesal60 asks whether
this chronological mismatch could be due to the con-
tinued use of the stamp dies, i.e. those of Augusti duo
(AVG[[G]]GNN[[[N]]) in the year 221 A.D., when
Heliogabalus adopted Severus Alexander, and Augustus
noster (AVG[[GG]]N[[NNN]]) under Alexander Severus
(222-235 A.D.). However, this hypothesis fails to take into
account one basic detail: that the dies were made of clay, as
56
Remesal Rodríguez 1996.
57
Berni Millet 2008, 201.
58
Testaccio 2001, nos. 153 and 443d.
59
Testaccio 2001, no. 443b.
60
Testaccio 2001, 260; Testaccio 2010a, 233.
BEZECZKY_Text.indd 142
Fig. 10.24 Titulus δ and the stamp on a Dressel 20 amphora manufac-
tured during the time of Caracalla (212-217 A.D.) and used
in the year 224 A.D. (Testaccio 2001, nos. 153 and 443d).
the letters A and N were removed when Severus, and later
Geta, died. The survival of this type of clay die for so many
years without repair, wear or breakage is hard to believe, as
making a new punch would have been quick and easy. It
seems strange that the dies for Augusti duo and Augustus nos-
ter would have been made after the death of Caracalla with
those modifications in the cancellation of the letters that had
ceased to have historical meaning.
The same dynamic, although with shorter time lapses,
was seen in the epigraphy of the finds from the 1998 exca-
vations on the Severian platform at Monte Testaccio. In that
case, three examples of AVG[[G]]GNN[[[N]] and four of
AVG[[GG]]N[[NNN]] were found together in strata attrib-
utable to the years 214 and 216 A.D.61 The stamps from the
Caracalla series were dated from the tituli picti to within the
historical period of that emperor (212-217 A.D.). By con-
trast, there was a chronological difference of between two
and four years in the Geta and Caracalla series.
In our opinion, the evidence from the stamps and graffiti
gives credibility to the hypothesis of the stockage of con-
tainers in the Baetican figlinae warehouses. We can confirm
that amphorae that had been manufactured during different
phases of the pottery’s activity were used in the diffusio olearia.
Some vessels were made and filled during the same season of
the year; others were manufactured in the autumn and taken
from the warehouse some months later. However, thanks to
the Testaccio finds, we also know that the ratio of stock in
the warehouses was much higher than that needed to meet
market requirements. The time that elapsed between the date
the amphora was manufactured and the date it was filled and
shipped is variable and unpredictable. It is difficult for us to
know the proportion in which these time lapses occurred as
we lack epigraphy (stamps and ante cocturam graffiti) that can
be compared to the consular dates from Monte Testaccio.
61
Testaccio 2010b, no. 314.
18.03.19 14:40
 CHAPTER 10 – Calendar graffiti on Dressel 20 amphorae (Berni Millet)
As a final conclusion, we have to insist once again on the
need to redefine the study of the Baetican stamps when it
comes to placing them in chronological order by the con-
textual dating from Monte Testaccio. Each stamp represents
a link in the evolutionary chain of the figlina activity. How-
ever, the excavations in Monte Testaccio show that it is risky
to establish seriations of families of stamps with a time span
of less than ten years. We can, however, classify them with
certainty in a time line with chronological leaps greater than
this safety margin.
62
Testaccio 2010a, 231.
BEZECZKY_Text.indd 143
143
The problem of stockage is extremely interesting as there
is no fixed time between each event. The mismatches in the
dating found at Monte Testaccio are unpredictable, with the
longest being found in the third century A.D. This is why at
Monte Testaccio there are overlaps in groups of stamps that
represent different and continuous evolutionary processes.
This is not a strange productive phenomenon matching in
time and of no use to understanding, as Remesal has recently
suggested with the new data from the Severian stamps from
Monte Testaccio.62
18.03.19 14:40
 144 EPIGRAPHY

7 UPDATED CHART OF CALENDAR GRAFFITI IN ORDER OF THE DATE OF THE YEAR

Date Graffito Origin (Dating) Bibliography
7 January Asiatici / vii Id(us) ian(uarias) Amat( )63 Reims (Marne, France) Photograph courtesy of Xavier Deru
10 January [M]artinus / iiii idus ianuarias / Gallo64 Testaccio Solin 1966-67; Rodríguez Almeida
1991, 36
13 January idibus ianuar[ii]s / Fortu[natus] Geneva (Switzerland) Paunier 1981, 243, no. 485
20 February Asiatici / sec( ) x k(alendas) marti(as) Alb( ) Brijuni (Croatia) (second century A.D.) Unpublished
February or [---]+++ / [---] martias / [---]us Testaccio (mid-second century A.D.) Rodríguez Almeida 1991, 37, Fig. 4B
March
February or [---]+[---] / [--- ma]rtia[s] / [---]ls+[---] Vindolanda (Chesterholm, Bardon Mill, Tomlin 2009, no. 70
March England)
16 or 26 March [---?] / [x]vii ‘aut’ vii k(alendas) april(es) / [---?] Testaccio (without context) Testaccio 1999, 26 Pl. XVIII, 68 no. 25
16-31 March [Ma]rtin[us] / [---] k(alendas) apriles / Meridi(anus) Testaccio (mid-second century A.D.) Rodríguez Almeida 1991, 36, Fig. 1;
Rodríguez Almeida 2000, Fig. 11
March or April [Orfito et Pri]sco cos / [---] k(alendas) apr(iles) Lenzburg (Switzerland) Martin-Kilcher 1987, Fig. 40.2
149 A.D. Barbari
3 April Prospectus? / iii n(onas) apriles / Vernaclus Velzeke (Belgium) Monsieur 2007, 104-105
7 April vii id(us) apr(iles) London (England) (context mid-first RIB 2493.2
century A.D.)
14-30 April Martini / [---] k(alendas) maias65 Testaccio (ca. 138-145 A.D.) Testaccio 2003, 374 no. 967, 387 Fig. 49
21 May Martinus / xii k(alendas) iunias / Tuscianus Testaccio (mid-second century A.D.) Rodríguez Almeida 2000, Fig. 11
27 May Martinus / vi k(alendas) iunias / Tuscianus Testaccio (mid-second century A.D.) Rodríguez Almeida 1991, 37, Fig. 2
30 May 153 A.D. Praesente et Rufin[o co]s / iii k(alendas) iun(ias) Augst (Switzerland) Rodríguez Almeida 1984, 256,
Barbari Fig. 91.3; Martin-Kilcher 1987, 71,
Fig. 40, Pl. 86 (1709)
May or June [---]i kal(endas) iu[nias, -lias] Vindolanda (Chesterholm, Bardon Mill, Tomlin 2009, no. 71
England)
May or June [Epap]hroditus / [--- i]unias / [S]ecundus66 Xanten (Germany) Weiss-König 2011, 231 no. 846
May or June [Me]lander / [--- iu]nias / [---]ilis67 Testaccio CIL, XV. 3609
16 May or 15 [---?] / xvii k(alendas) iu[nias] ‹aut› iu[lias] / +[---] Testaccio CIL, XV. 3606
June
26 May or 25 vii k(alendas) iu[nias] ‹aut› iu[lias] / Sosimu[s]68 Birrens (Dumfriesshire, Scotland) RIB 2493.4
June
18 June Menander / xiiii k(alendas) iulias ‘Thermenmuseum’, Heerlen (Holland) Van Hommerich 1966, 10.
Rodríguez Almeida 1993, Fig. 3.2;
Photograph courtesy of Radboud
University, Nijmegen (Department of
Roman Provincial Archaeology)
23 June 158 A.D. Tertullo et Sa[cerdote cos] / viiii k(alendas) iul(ias) ‘De Woerd’ Valkenburg (Holland)69 Rodríguez Almeida 1981, 123,
Lucrio[nis] Fig. 11.5; Rodríguez Almeida 1984,
256, Fig. 2. Moros – Berni Millet
2011, 44, Fig. 13
1 July Masiedi? / k(alendis) iuliis70 Testaccio (ca. 138-145 A.D.) Testaccio 2003, 374 no. 957, 386 Fig. 48
6 July ii non(as) iuli[as] / +[---]71 Mâcon (Saône-et-Loire, France) (late Liou 1987, 136 (MAC 10), Fig. 41
first - early second century A.D.)
20 July (cited, without sketch or text) Testaccio (mid-second century A.D.) Rodríguez Almeida 1984, 256 note 6
(B-7/ 1818)
22 July [---?] / xi k(alendas) aug[ustas] / [---?]72 Carlisle (England) (Trajan context) RIB 2493.3
25 July Asiatici / viii k(alendas) au[g(ustas) ---]73 Brough on Noe (Derbyshire, England) Wright – Wilson 1963, 166 no. 50;
RIB 2493.46
July or August [---?] / [--- a]ugustas / [---?] Testaccio (ca. 138-145 A.D.) Testaccio 2003, 375 no. 972, 388 Fig. 50
July or August [---?] augus(tas) / [---]+74 Testaccio CIL, XV. 3607
July or August [---]troddus / [--- ]aug(ustas) / [---]+ Testaccio (mid-second century A.D.) Rodríguez Almeida 1991, 37, Fig. 5.B
19 August Martinus / xiiii k(alendas) sep(tembres) / Gallo75 Testaccio (context 153-154 A.D.) Rodríguez Almeida 1991, 40, Fig. 6;
Testaccio 1999, 27 Pl. XIX, 68 no. 24
20 August Asiatici / xiii k(alendas) sep(tembres) Luc( )76 Lincoln (England) Wright 1960, 221, 241 no. 35; RIB
2493.6, Pl. VI
26 August +dsu++[---] / vii k(alendas) sept(embres) / Cumberland (England) RIB 2493.7
In+o+s+++[---]77
15 September Condiano cos / xviii k(alendas) octobr(es) Catonis? Torino (Italy) Rodríguez Almeida 1984, 256
151 A.D. note 7, Fig. 91.4

BEZECZKY_Text.indd 144 18.03.19 14:40

 CHAPTER 10 – Calendar graffiti on Dressel 20 amphorae (Berni Millet) 145

Date Graffito Origin (Dating) Bibliography

19 September xiii k(alendas) octobres “De Holdeurn”, Nijmegen (Holland) Holwerda – Braat 1946, Pl. XXV, no.
2031. Photograph courtesy of Radboud
University, Nijmegen (Department of
Roman Provincial Archaeology)
29 September Melande[r] / iii k(alendas) octo[bres] / Gallo Adelfa (Seville, Spain) Moros – Berni Millet 2011, 41
29 September iii k(alendas) octobres / Fortunatus Geneva (Switzerland) Paunier 1981, 243, no. 484
3 October Menandri / v nonas octobr(es) Tiel (Holland) Photograph and sketch courtesy
of Radboud University, Nijmegen
(Department of Roman Provincial
Archaeology)
20 October [---?] / xiii k(alendas) nov[embres] / Caesi[---] Catterick (England) Burnham et al. 1997, no. 31
21 October [---?] xii k(alendas) n[ovembres] [---?]78 London (England) Tomlin 2009, no. 43
October or [---] nove[mbri(bus, -es)] Lincolnshire (England) RIB 2493.11
November
October or [---] novembr[i(bus, -es)] [--- ?] Bonn (Germany) Bakker et al. 1975, 169 no. 553
November
1 November kal(endis) novembri(bus) “prope castellum Hofheim” (Germany) CIL, XIII. 10003.16
1 November <kalendis> novembri(bus) Digoin (France) Maza et al. 2014, 823, Fig. 46
16 December, 16 Asiatici / xvii k(alendas) i[an(uarias), -un(ias), Wall, Staffordshire (England) Taylor – Collingwood 1925, 250
May, 15 June -ul(ias)] [---]79 no. 22; RIB 2493.5
28 December [---?] / v k(alendas) ia[nuarias] / Heli( ) Testaccio CIL, XV. 3606
28 December Asia[tici] / v k(alendas) ian[uarias ----] Vindolanda, Bardon Mill (England) Courtesy Andrew and Barbara Birley
29 December, 29 Asi[atici] / sec( ) iiii k(alendas) i[an(uarias), Altenstadt (Germany) Schönberger – Simon 1983, 161 no.
May, 28 June -un(ias), -ul(ias)] [---]80 55, Pl. 68.D55; Ehmig 2007, Pl. 1.169
30 December (cited without sketch or text) Testaccio (mid-second century A.D.) Rodríguez Almeida 1984, 256 Note 8
(B-7, 8/1786)

63
See Section 5.
64
According to Solin: “[---]minus / sec(undo) idus Ianuarias / Gallo”.
65
According to the publication: “[---] / Martini / [---]maius” (“the
remains of a possible calendar graffito”).
66
Complete the first name as [Epap]hroditus instead [Ap]hroditus.
67
Dressel’s original reading is “[--- Ale]xander / [--- iu]nias / [---apr]
ilis”. Rodríguez Almeida (1990, 37) reads the first name as [---]
nder and suggests completing it as Menander, Alexander, etc. Both
suggestions should now be corrected as Melander thanks to the
find from Adelfa with the date 29 September (see general chart).
Another detail to be corrected is in the word [apr]ilis on the last
line, if Dressel understood it as the fourth month of the year, as
the inclusion of a second element of the Roman calendar does not
fit with the known formulas of Baetican calendar graffiti. This last
word is linked to the subordinate person and is probably written in
the nominative case: [Vir]ilis, [Ag]ilis, or even [Apr]ilis as a proper
name.
68
Read as “vii k(alendas) iu[n(ias)] or iu[l(ias)] / [---]cosimu[---]?”.
69
Rodríguez Almeida published it several times as a find from Nijmegen,
acknowledging José Remesal for the first report of it and subsequently
J.E. Bogaers of the Catholic University of Nijmegen for having sent
him the university’s photograph and tracing (Rodríguez Almeida
1981, 123). In reality, thanks to our colleague and friend Joost van
den Berg, we know that the Dutch find comes from the “De Woerd”
archaeological site in the town of Valkenburg and is deposited in the
Bureau Archeologie en Monumenten – Nijmegen, where we saw and
photographed it.
70
Published as “Masiedi / k(alendas) eprilis” (“spelling mistake: eprilis
instead of aprilis”). The absence of the calculation of the days pre-
ceding the calendas is because we are on the first day of the month,
perhaps July, judging by the length of the strokes in the sketch and the
ending in “iis” of the plural ablative.
71
Perhaps a two-line calendar graffito. In the published photograph we
can clearly see the curved, raised stroke of the first letter of the pos-
sible cognomen on the second line.
72
Published as “[---?] / [---]xv? k(alendas) ma[rtias] or ma[ias]”. Based
on the sketch, we believe the number should be xi instead of xvi and
the name of the month should be aug[ustas].
73
See Section 5.
74
Published as “[---] augus”. We cannot rule out the possibility that this
is a binominal graffito “Augus(talis) / [---]+”.
75
We have to point out two irregularities in Rodríguez Almeida’s read-
ing: “Martinus / xiii k(alendas) sep(tembres) / Gallio?”. Firstly, the
Fig. XIII on the calendar date, as in both the photo and the sketch
we can clearly make out the fourth stroke of the unit of fourteen days
(Fig. 5). Next, the supposed nexus “io” in the termination of the cog-
nomen “Gallio?”, as in the sketch we can also see clearly how the last
two strokes form the typical letter ‘o’, albeit with the second stroke
elongated downwards to make it the last letter of the word and the
text.
76
See Section 5.
77
Published with doubts as: “Suar[---] / vii K(alendas) sept[embres] /
Insolsar[---]”.
78
Published as: “[---]xii k(alendas) m[aias, -artias]”. Judging from the
sketch of the graffito, the first letter of the month appears to us to be
an ‘n’.
79
See Section 5.
80
See Section 5.

BEZECZKY_Text.indd 145 18.03.19 14:40

BEZECZKY_Text.indd 146 18.03.19 14:40

11 INSCRIPTIONS ON LATERITE FINDS FROM BRIJUNI
Piero Berni Millet
The Brijuni National Park Archaeological Museum
(BNPAM) has a splendid and very interesting collection
of non-amphora finds with epigraphy on laterite (tegula,
imbrex and dolium). In this paper we will explain these finds,
an assemblage of 131 pieces, the vast majority of which are
stamps on tegulae imported from various manufacturing
regions in the Upper Adriatic area.1
The objective of this paper is to offer an updated catalogue
of the Brijuni laterite stamps with the new photographic
documentation prepared by Bezeczky, which will help to
improve our understanding of the records of this type of
inscription. We include much new information that was
not in the previous epigraphic catalogue of tegulae stamps
from the southern coast of Istria compiled by Matijašić pub-
lished in 1987. Moreover, we had the advantage of being
able to approach the study of the northern Adriatic stamps
from a more up-to-date research perspective, as consider-
able advances have been made in the last twenty years. It is
not our intention to undertake an in-depth and exhaustive
analysis of these inscriptions (meanings, onomastics, pro-
duction, markets, etc.), but to report on the finds with the
basic epigraphic study information.
The majority of the Brijuni laterite stamps come from
three of the island’s best known archaeological sites: the luxu-
rious Roman seaside villa of Val Catena (Verige), the Roman
villa of Madona bay – which was subsequently absorbed by
the walled enclosure of the Byzantine castrum – and the
olive oil producing villa/factory of Monte Collisi (Kolci).2
The earliest finds are linked to Gnirs’ archaeological excava-
tions in the early twentieth century.3 They are mainly from
the Roman villa of Val Catena, with one from Monte Col-
1
We would like to thank Mira Pavletić, director of the Brijuni Museum,
for allowing us access to these important finds to carry out this study.
We would also like to express our most sincere gratitude to Tamás
Bezeczky for giving us the opportunity to participate in his Brijuni
project for the cataloguing and analysis of these finds, a task we faced
for the first time in the hope of complying with the scientific demands
required by the subject.
2
See La Torre – Bezeczky in this volume.
3
Gnirs 1904; Gnirs 1908; Gnirs 1910.
BEZECZKY_Text.indd 147
147
lisi and another from the Basilica of Val Madona. The oth-
ers are from more recent excavations, including those carried
out in the Castrum between 1976 and 1980,4 some of which
were deposited at the time in the Pula Archaeological Muse-
um.5 Matijašić’s catalogue contains an important assemblage
of stamps from the Castrum with the letters BA. We have
seen and re-documented these inscriptions and added a few
previously unpublished pieces recorded with those initials,
which we attribute by default to the finds from the Cas-
trum. Also from the Castrum are seven stamps published by
Vitasović in 2007 with the finds from the 1976 and 1984
excavations. The museum also has some stamps from exca-
vations carried out in the 1970s, including the six from the
1979 excavation of the lower zone of the Roman villa of Val
Catena published by Vitasović (2004). Among the new pho-
tographic documentation in this catalogue there are many
unpublished inscriptions with no clear indication of where
they were found on Brijuni Island and which could therefore
have come from any of the aforementioned archaeological
sites.
In order to provide a record of all the laterite finds with
stamps in the Brijuni Museum, in this catalogue we also
include a few foreign pieces brought to the island from Pula,
with references to excavations in the 1930s and 40s and asso-
ciated with outstanding Roman urban structures of a public
and monumental nature (amphitheatre, theatre, mausoleum,
etc.). The following is a summary of the quantitative informa-
tion by origin of the finds: 44 from the Castrum, 16 from the
Roman villa of Val Catena, 2 from Monte Colisi, 1 from the
Basilica of Val Madona, 6 from Pula and 62 with an unde-
4
Matijašić 1987, note 8; Mlakar 1975-76.
5
The Archaeological Museum of Istria in Pula has an important collec-
tion of stamped tegulae fragments brought from all over the península
and calculated by Matijašić to be some 800 pieces (Matijašić 1985;
Matijašić 1993b, 131). The majority of these finds are from pre-
1947 excavations or are chance finds for which the museum has no
documentation. Among the few exceptions Matijašić refers to are the
stamps from Gnirs’ Brijuni excavations. There are other stamps from
Brijuni from more recent excavations, although we have no details of
them.
18.03.19 14:40
 148 EPIGRAPHY
termined origin. All these inscriptions with their respective
initials and origins are listed in a summary at the end of the
corpus in the general epigraphic index in Section C.
The stamps that can be dated from parallel finds are lim-
ited to a well-defined chronological horizon between the
mid-first century B.C. and the second half of the first century
A.D., approximately between the late Republican Caesarean
period and the reign of Domitian. Thus we have epigraphic
evidence of building materials being brought to the island in
different periods that can be attributed to different construc-
tive and evolutionary phases of these settlements.
The regions that produced stamped laterite are quite
numerous and spread across the arc of the northern Adriatic
coastline. We can definitely single out three large zones that
supplied building materials, two of them foreign, on the Ital-
ian peninsula, and one local, on the western Istrian coast.
Firstly we should mention the 67 items from eastern Cis-
alpine Gaul with a well-defined chronological horizon from
the mid-first century B.C. to the time of Nero. The figlina
Faesoniana, whose production centre was between the Ital-
ian towns of Rimini and Pesaro, provides the largest number
of stamps, with 44 from the A·FAESONI·AF series (cf. no.
7). It is followed by the figlina Pansiana in the town of Fer-
rara with 22 stamps that can be divided into five large evo-
lutionary groups representing the pottery’s important and
long-lasting industrial activity (cf. no. 12). The Cisalpine
Gaul list is completed by 2 C·CEIONI·MAXI stamps, a
production believed to have come from the towns of Rimini
and Ravenna.
The second major exporting zone was in the Italian region
of Friuli-Venezia Giulia, in the area of the provinces of Udine
and Trieste. Standing out over all of them are the laterite pro-
ductions from the territory of Aquilea, including those of the
well-known entrepreneur Q. Clodius Ambrosius, the largest
private manufacturer of laterite in the first century A.D., of
whom we have 5 documented examples on tegulae found in
the Castrum (cf. no. 6). From the Petronii of Aquilea we have
3 stamps represented by the LPETRAVIT and LPETRO
series. The first is dated to the first century A.D., while the
other is not found in the epigraphic corpora and could be
dated, judging by the onomastic formula of the duo nomina,
to an earlier manufacturing phase (cf. no. 13). Other minor-
ity evidence from this zone can be dated to the mid-first cen-
tury A.D.; for example, C·IVLI·AFRICANI (cf. no. 10) and
L·ST·IVSTI (cf. no. 16), both found on a single example. We
also have 3 TROSI stamps, one on imbrex dated between the
late first century B.C. and the early first century A.D. The
gens Trosia enterprise appears to have been a decentralised
BEZECZKY_Text.indd 148
manufacturing operation in the Friuli region, with vari-
ous potteries dotted around the ancient Roman territorial
demarcations of Aquileia and Tergeste (cf. no. 17). We do
not wish to conclude this section without briefly referring to
two interesting pieces that were found out of their archaeo-
logical context on Brijuni Island and are also studied in the
catalogue. The first inscription, L·BARBI·L·L·EVPOR[IS]
& L·BARBI·L·L·EVPOR[I], is impressed on a fragment of
imbrex found in the Pula amphitheatre (cf. no. 3). The per-
son referred to in the stamp was a freedman of the Barbii
of Aquilea, one of the most important families among the
first century B.C. laterite owners and producers, of which
there are numerous members of different social statuses
recorded on laterite. The second is a late Augustan stamp,
M·ALBI·RVFI, from the Campidoglio Theatre in Pula (cf.
no. 1), from a production centre possibly in the territory on
the border between the Veneto and Friuli regions.
At this point it is desirable to pause for a moment to men-
tion some of the northern Adriatic laterite productions of
undetermined origin. Of particular interest for the number
of finds made is the M·SERI series with 10 examples, four
associated with the Roman villa of Val Catena and one defi-
nitely from the Castrum (cf. no. 14). An important detail to
bear in mind is that this stamp has only been documented in
Brijuni and Carnizza/Krnica; we know of no parallels out-
side Istria. Therefore, we have to ask ourselves if the M. Serius
tegulae were actually manufactured in Istria, given the origin
of all the finds, or whether they came from outside, based on
the nationality of the gentilic Serius, which is widely found,
for example, in stamps on Italian terra sigillata from the Po
Valley. Still the subject of controversy is the origin of the
C. Flavius laterite stamps found in Istria, with 7 examples
in Brijuni (C·FLAVI and C·FLAV), two of them definitely
from Val Catena (cf. no. 8). This person has been linked to
the mark C·FLAVI on two Dressel 6B amphorae, one from
Padua and the other from Monselice, which leaves open the
possibility of a non-Istrian origin, perhaps also in the Po
Valley, although other members of the gens Flavia recorded
on laterite are from different geographical locations, in the
southeast of the province of Venice (C·FLAVIF) and the
Emilia-Romagna region (L·FLAV/PHILIRI).
The following controversial case concerns the
L·FVLLONI stamp (cf. no. 9) of which we have 7 examples,
two definitely from the Castrum and another from Val Cat-
ena. At present we have little information about this laterite
manufacturer from around the turn of the era. Two possible
production zones are currently in play: Friulian, from the
evidence documented in Aquilea, or perhaps Istrian, from
18.03.19 14:40
 CHAPTER 11 – Inscriptions on laterite finds from Brijuni (Berni Millet)
the reddish-brown colouring of the ceramic paste. Of greater
difficulty is the study of the Q·G·NICEP (cf. no. 4) stamp,
as the only known examples are the 5 from Brijuni Island.
We have asked ourselves if the initial letter G( ) of the gen-
tilic could represent a freedman of the gens Grania, a laterite
manufacturing family from Iulia Concordia, an Italian local-
ity in the province of Venice bordering on the Friuli region.
Finally, we have to mention the fragmented inscription
ARIAL[---] from the Castrum (cf. no. 18), the only stamp
documented on dolium, probably referring to a woman, per-
haps linked to the gens Ar(r)ia and found on other types of
building material in Trieste and northeastern Istria from the
early Roman period in the first century A.D.
The third and final confirmed manufacturing region is on
the western coast of Istria. This was the home of the Fažana
and Loron figlinae, two of the largest olive oil amphora man-
ufacturing centres in Istria with a varied, mixed and diver-
sified ceramic industry, which also manufactured building
materials for regional use in towns and villae. These two
potteries came to have parallel lives during the first century
A.D., initially as the domains of rich owners from the sena-
torial order and later as Imperial property under the Flavian
dynasty.
The large amphora factory in Loron was part of an exten-
sive seaside complex installed around the two bays of Cer-
var and Santa Marina, to the north of Parentium (Parenzo/
Poreč), one of the three main urban centres of Roman His-
tria; the other two were Pola and the frontier town of Ter-
geste. It was a pleasant surprise to come across the SISEN-
NAE stamp – which is well-known from Loron olive oil
amphorae – for the first time on a fragment of tegula held in
the Brijuni Museum (cf. no. 15). Hitherto we had only seen
it, in an identical form, on the rims of Dressel 6B ampho-
rae and on the bases of terra sigillata vessels produced in the
local figlina. The stamp records the proprietor T. Sisenna
Statilius Taurus, who was an ordinary consul in the year 16
A.D. The industry may have been founded at the time of his
grandfather, the great T. Statilius Taurus, a friend of Augus-
tus and ordinary consul in 37 and 26 B.C. From the Imperial
production phase of the Loron pottery we have a probable
documented stamp, IMP·AVG·GER (cf. no. 2), attributable
to Domitian, who took the name Germanicus from the year
83 A.D.
The large Fažana amphora factory was on the southern
coast of Istria, facing Brijuni Island and 8 km northwest
of Pula. In the Roman period the figlina and its surround-
ing territory were part of the ager of Pola. From the second
half of the first century B.C. until the arrival of the Flavian
BEZECZKY_Text.indd 149
149
dynasty the pottery belonged to the Caii Laecanii Bassi, who
boasted two generations of consuls. The Laecanii are con-
sidered to have been the largest producers of Istrian olive oil
and their amphorae supplied the markets of northern Italy
and the Alpo-Danubian provinces (Noricum, Reatia and
Pannonia). This family also owned Brijuni Island with its
three villae, including the sumptuous palace of Val Catena, a
fact that is corroborated by the laterite epigraphy discovered
on the island. This catalogue includes an important assem-
blage of 14 Laecanii stamps on tegulae (cf. no. 11), six defi-
nitely from the Castrum and one from Val Catena. The only
example of the C·LAECANI·P·F series is believed by various
experts to be one of the first productions of the figlina. The
others, with the mark LAEC·, are classified with four varia-
tions and are dated, possibly, to a subsequent late Augustan
phase (ca. 10 B.C.-14 A.D.).
In conclusion, we will briefly recall what was behind
this interesting trade in building materials of such diverse
origins during a well-defined historical period between the
Late Republic and the first half of the Early Roman Period.
It was an economic and social phenomenon that affected
the entire Adriatic arc, from the Italian region of Marche
to the coasts of Dalmatia, and generated large-scale trade
between the principal economies and the emerging regions.
In the Roman territories of eastern Venetia and Histria this
phenomenon was linked to the founding and development
of its main towns, rural settlements and manufacturing
complexes, which received an important and decisive eco-
nomic boost from the mid-first century B.C., following the
end of the Second Civil War between Caesar and Pompey.
Construction intensified with the monumentalisation of
the administrative centres of the main towns: Iulia Con-
cordia, Aquileia, Iulium Carnicum, Forum Iulii, Tergeste,
Parentium, Pola and Nesactium. This geographical area
constituted an economic unit that can be analysed through
the production and circulation of stamped laterite, from
which we can see a widespread and intensive trade in build-
ing materials at that time. This whole region, like the rest of
Gallia Cisalpina, was particularly rich in fired industrial-type
building materials: tegula, imbrices, lateres, antefix, tubuli,
etc. The majority of the stamped items are rectangular and
flat tegulae (60 × 42-45 cm). There is little epigraphic evi-
dence on imbrices and lateres. Tegulae were not only used as
roofing material, but were often also cut into pieces to use
in building foundations,6 for example, as a facing on opus
caementicium walls, a construction technique already doc-
6
Zaccaria – Gomozel 2000, 288.
18.03.19 14:40
 150 EPIGRAPHY
umented in the time of Caesar and Augustus that became
widespread during the reign of Tiberius.7
To end this general introduction, we will briefly comment
on the methodology used in drawing up the epigraphic cor-
pus, so that the reader can better understand its organisa-
tional structure. The catalogue has been divided into two
sections for the purpose of distinguishing the abundant
epigraphy on laterite (tegulae and imbrices) from the few
examples on dolia (one stamp and a graffito denoting capac-
ity). In the main section, the inscriptions are classified using
the nominal system, with the aim of making it easier to link
families of stamps. The catalogue is structured in accordance
with the following criteria: the main numbering identifies
a family of stamps, while the corresponding letters indi-
cate the different variations and the numbers the examples
of each variation. The meanings of the abbreviations in the
descriptive fields are as follows: MP (place of manufacture),
Dating (dating), PF (place of find), PC (place of conserva-
tion), Reading (reading of the stamp), Lit. (literature, if the
inscription has been previously published), Desc. (addi-
tional observations on the object). The initials BNPAM in
the PC field refer to the Brijuni National Park Archaeologi-
cal Museum. When the Latin key word ‘vidimus’ appears
next to the Lit. field, it means we have seen the piece.
A. EPIGRAPHY ON LATERITE (TEGULAE,
IMBRICES)
1. MALBIRVFI
PM: Veneto-Friuli
Dating: ca. 10-15 A.D.
a) [M·ALBI·R]VFI (Plate 23)
PF: Pula, “Theatre of the Campidoglio”
PC: BNPAM, A 10643
Reading: [M. Albi R]ufi
Lit.: Vidimus. Matijašić 1987, 40
The only example of M·ALBI·RVFI in the Brijuni
Museum comes from Pula. In its original state this stamp had
three linked groups: A^L, B^I and V^F, letters with incised
relief and no frame. In his catalogue Matijašić includes more
Istrian examples with better preserved text; two are chance
finds from Barbariga and the third is from Pula, which he
attributes to Gnirs’ excavations in the early 20th century.8 We
7
Righini 1998, 54.
8
Matijašić 1987, nos. 55, 133-134.
BEZECZKY_Text.indd 150
also have to add to this list the stamp in the Pula Museum
catalogued in CIL, V. 8110.36c.
The possible origin of this production can be found in the
Italian regions of Veneto and Friuli. The finds are spread over
the territories of Tergeste,9 Aquilea10 and Concordia Sagit-
taria (Iulia Concordia).11
A dolium found in the locality of Dossi (southeastern
Aquilea) preserves two quite worn stamps on which we can
read M·ALBI·RVFI, which would confirm another case of
mixed pottery production12. The M·ALBI·RVFI stamp was
part of a laterite cargo found in a sunken ship in the River
Stella (Pescarola) dated to ca. 10-15 A.D.13
We can see a family link between the laterite produc-
tion of M. Albius Rufus and the person referred to in the
M·ALBI·MACRI stamp14, which can also be dated to the
first century A.D., based on the onomastic elements and the
regional evidence of the finds in the ager aquileiensis.15 It is
considered to be one of the most widespread stamps in the
whole of the Adriatic arc, an area that stretches from the cen-
tral Italian region of Marche to Dalmatia.
2. IMPAVGGER
PM: Loron? (Istria)
Dating: post quem 83 A.D.
a) IMP·AVG·G[ER] (Plate 23)
PF: -
PC: BNPAM, s/n
Reading: Imp(eratoris) Aug(usti) G[er(manici)]
Lit.: Vidimus.
A single, poorly impressed example with thin-bod-
ied letters in relief. The inscription records the formula
IMP·AVG·GER, which provides a post quem dating from
the title “Germanicus”, received by the emperor Domitian in
83 A.D. to commemorate his military victories in Germania.
This production is from Istria, but there is some doubt as to
whether to attribute it to Fažana or Loron, given that both
figlinae were under Imperial ownership and they both made
and stamped amphorae and building materials in the time of
Domitian.
9
Zaccaria – Zupančić 1993, 139 no. 2.
10
CIL, V. 8110.36a-b, d-e.
11
G.A.V.O. 1993, 208.
12
Bertacchi 1992, 21-24; Zaccaria – Gomezel 2000, 299, note 87.
13
Zaccaria – Gomezel 2000, 299 and 301.
14
CIL, V. 8110.35.
15
Gomezel 1996, 39 and 58.
18.03.19 14:40
 CHAPTER 11 – Inscriptions on laterite finds from Brijuni (Berni Millet)
The Imperial stamps of Domitian on tegulae and Dres-
sel 6B amphorae have a confirmed origin in Loron. Various
stamps on tegulae were found during the excavations in the
kiln sector of the pottery.16 Moreover, the same formula
IMP·AVG·GER is found impressed on the rims of Dres-
sel 6B amphorae.17 Also from Loron there is another varia-
tion on amphorae from this emperor with the formula IMP
((hedera)) DOM ((ramus palmae)).18 To date in Fažana no
stamps on amphorae have been found that use the imperial
formula with the name Germanicus, although there are other
variations of the IMP·DOMITIANI or IMP·DOM type
from the same emperor.19
3. LBARBILLEVPOR[IS] &
LBARBILLEVPOR[I]
PM: Aquilea
Dating: first century B.C.
a) L·BARBI·L·L·EVPOR[IS] &
L·BARBI·L·L·EVPOR[I] (Plate 23)
PF: Pula, amphitheatre “maggio, 1938”
PC: BNPAM, A 10628
Reading: L. Barbi L. l(iberti) Eupor[is] aut Eupor[i]
Lit.: Vidimus. Matijašić 1987, no. 49
Desc.: Imbrex. Size 19 × 13.5 × 2.5 cm; stamp 7.8 cm;
letter height 1.2 cm.
A retrograde inscription stamped on a slightly convex
fragment of yellowish imbrex from the Pula amphitheatre.
The object was impressed from left to right, with less pres-
sure at the end, meaning that the last letters can only be read
partially and with some difficulty. Matijašić sketched and
catalogued it with the reading LBARBILLEVP, without
punctuation marks. Nevertheless, in the photograph we can
clearly see the cognomen, as well as the detail of the four thin,
rounded interpunctuation marks that separate the five words
identifying the person in the inscription.
The stamps of the freedman of L. Barbius are quite well
known on laterite thanks to other sets of dies of varying sizes
and composition – L·BARBI·L·L·EVP and L·BARBI·L·L·EV,
L·BARBI·L·L – that, when used, produced thin-bodied let-
ters in relief written from left to right. Our variation on
imbrex records, for the first time, the person’s full cognomen.
16
D’Incà et al. 2010, 325.
17
Marion – Starac 2001, 107-109.
18
Tassaux 2010.
19
Tassaux 2010.
BEZECZKY_Text.indd 151
151
Fig. 11.1 The final letters EVPOR visible in the retrograde
stamp on imbrex (Photo: Bezeczky 2014).
The faintness of the impression at the end of the text makes
it difficult to read the outline of the two letters that follow
EVP with absolute clarity. Moreover, neither can we see the
left-hand closure of the frame which is so necessary for defin-
ing the end of the inscription. In our opinion the shadows
of the last two visible letters are OR and therefore we agree
with Zaccaria – Zupančić20 when they suggest a read-
ing of Eup(oris), although we cannot rule out the alternative
reading of Eupori.
The Barbii from Aquilea was one of the most rep-
resentative families among the first century B.C. later-
ite owners and manufacturers, with numerous members
belonging to different social classes.21 In the first phase of
laterite production we find various ingenui with different
praenomina: L·BARBI·L·F, L·A·BARBI·L·F, L·BARBI·TI·F,
Q·BARBI·Q·F, T·BARBI·L·F, T·BARBI·TI·F. In this initial
stage the freedman Eupor or Euporus is consolidated with
various sets of stamps impressed on various types of pot-
tery (tegulae, lateres, imbrices, etc.). Later, during a period
of expansion of manufacturing in the Cividale area, we find
other stamps with a new type of formula that records the
names of slaves: AGATOCL·BARB or BARB·AGATOCL,
TIRO·BARB or BARB·TIRO. Perhaps they managed the
laterite factories belonging to the gens Barbia at a time of
growth related to the urban development of Forum Iulii.22
The Barbii are also present in the manufacture of dolia with
the stamp T·BARBI·TI·F.23
The Barbii ’s artefacts were mainly distributed in the Ital-
ian regions of Veneto and Friuli. The majority of the finds
20
Zaccaria – Zupančić 1993, 139 no. 7.
21
Gomezel 1996, 95-96; Zaccaria – Gomezel 2000.
22
Gomezel 1996, 96.
23
Zaccaria 1991, 305-306; Zaccaria – Gomezel 2000, 299, note
87.
18.03.19 14:40
 152 EPIGRAPHY

come from the western field of the Roman town of Aquilea.24 Five stamps – all from the same die – with the text in pos-
They are also well-documented in the territory of Tergeste.25 itive relief in a rectangular frame. The letters are sharp and
Our person is known in Istria on laterite with the shorter full-bodied, although somewhat squashed together. The two
variations. He is found on a fragment of tegula excavated by punctuation marks are triangular and the text was resolved
Gnirs at “Porta Ercole”,26 as well as in the Roman town of with a single nexus, E^P.
Nabresina/Nabrežina (Slovenia)27 and the locality of San Since 1987, when Matijašić published the sketch of the
Lorenzo in Daila (Croatia).28 incomplete example (a3) with the transcription QGNIC[-
--] without punctuation marks, no subsequent studies have
4. QGNICEP attempted to restore the missing part of the inscription. For-
PM: Veneto? tunately we now have four new finds from Brijuni Island to
Dating: first century A.D.? help us complete the reading of this person’s cognomen.
We now know that the Q·G·NICEP stamp refers to
a1) Q·G·NICEP (Plate 23) a freedman called Nicep(horis) or Nicep(hori) from an
PF: - unknown family whose identity lies behind the initial G( )
PC: BNPAM, B 91.22 of a Roman gentilic. Could this initial represent a freed-
Reading: Q. G( ) Nicep(horis, -i) man of the gens Grania. One member of this family is the
Lit.: Vidimus. well-known laterite manufacturer, Q. Granius Priscus, from
the territory of Iulia Concordia in the province of Venice.29
a2) Q·G·NIC[EP] (Plate 23) Among his stamps we find the variations Q·GRANI·PRISCI
PF: - and Q·G·PRISCI,30 the latter with the lettering of the name
PC: BNPAM, B 91.1 similar to the case shown here.
Reading: Q. G( ) Nic[ep(horis, -i)] The possible dating of the Q·G·NICEP stamp con-
Lit.: Vidimus. cerned could be limited to the first century A.D., taking into
account the onomastic formula used with the tria nomina
a3) Q·G·NIC[EP] (Plate 23) for a freedman.31
PF: Castrum
PC: BNPAM, BA 1903 5. CCEIONIMAXI
Reading: Q. G( ) Nic[ep(horis, -i)] PM: Cisalpine Gaul
Lit.: Vidimus. Matijašić 1987, 26 Dating: ca. second half first century B.C. - mid-first
century A.D.
a4) [Q·]G·NICEP (Plate 23)
PF: - a1) C·CEIONI·MAXI (Plate 23)
PC: BNPAM, B 91.11 PF: Monte Collisi
Reading: [Q. ]G( ) Nicep(horis, -i) PC: -
Lit.: Vidimus. Reading: C. Ceioni Maxi(mi)
Lit.: Gnirs 1908, Fig. a
a5) [Q·]G·NICEP (Plate 23) Desc.: Stamp height 3.5 cm
PF: -
PC: BNPAM, B 91.15 a2) [C·]CEIONI[·]MAXI (Plate 23)
Reading: [Q. ]G( ) Nicep(horis, -i) PF: Castrum
Lit.: Vidimus. PC: -, BA 4130
Reading: [C.] Ceioni Maxi(mi)
Lit.: Matijašić 1987, 30
Desc.: Size 16.7 × 16.7 × 3 cm; stamp 12.5 × 3.4 cm;
24
Furlan 1993, 199; Cargnelutti 1992, 103; Cargnelutti
1993, 87.
25
Zaccaria – Zupančić 1993, 139 no. 6.
26
Matijašić 1987, no. 57. 29
Gomezel 1996, 99.
27
CIL, V. 8110.53a-c. 30
Furlan 1993, 199.
28
Zaccaria – Zupančić 1993, 139 no. 7. 31
Gomezel 1996, 39-44.

BEZECZKY_Text.indd 152 18.03.19 14:40

 CHAPTER 11 – Inscriptions on laterite finds from Brijuni (Berni Millet) 153

letter height 2.6 cm; nexus M^A, the letter X is a1) QCLODIAMB[ROSI] (Plate 23)
smaller and higher. PF: Castrum
PC: BNPAM, BA 9447
We were unable to find the only two examples of the Reading: Q. Clodi Amb[rosi]
C·CEIONI·MAXI series from the Castrum and the Monte Lit.: Vidimus. Matijašić 1987, 33
Collisi olive oil factory that are supposed to be in the Bri-
juni Museum. From Gnirs’ and Matijašić’s sketches we a2) QCLODIAM[BROSI] (Plate 24)
understand that there are two stamps from the same varia- PF: Castrum
tion, with the tria nomina separated by two small triangular PC: BNPAM, BA 1904
points and the cognomen MAXI abbreviated with a M^A Reading: Q. Clodi Am[brosi]
nexus, followed by a smaller, raised letter X. Lit.: Vidimus. Matijašić 1987, 35
We have little information about laterite productions
signed with the gentilic Ceionius. The origin of the produc- a3) QCLODI[AMBROSI] (Plate 24)
tion appears to be eastern Cisalpine Gaul, where stamps nam- PF: Pula, Mausoleum
ing various members of this family are documented quite fre- PC: BNPAM, A 10629
quently on tegula. On the laterite from the regions of Rimini Reading: Q. Clodi [Ambrosi]
and Ravenna there are some examples of the C·CEIONI Lit.: Vidimus. Matijašić 1987, 42
and C·CEIONI·MAXI series attributable to Caius Ceio-
nius Maximus.32 This person must have been related in a4) QCLO[DIAMBROSI] (Plate 24)
some way to the Lucius Ceionius of the tegulae from the ter- PF: Castrum
ritory of Bologna, whose stamps have the particularity of PC: BNPAM, BA 4135
recording the names of slaves with the patronymic formula Reading: Q. Clo[di Ambrosi]
HILARVS·PIL·CEION·L·S, HILARVS·VIR·CEION·L·S, Lit.: Vidimus. Matijašić 1987, 36
[E]ROS·CEION·L·S.33 It is therefore possible that from the
end of the Republican period to the time of the Julio-Clau- a5) [QC]LODIAMBRO[SI] (Plate 24)
dian dynasty L. Ceionus and C. Ceionius owned brick facto- PF: Castrum
ries in what is now the Italian region of Emilia-Romagna. PC: -
As far as the Istrian finds are concerned, we have exten- Reading: [Q. Cl]odi Ambro[si]
sive documentation within what can be termed a discreet Lit.: Vitasović 2007, Sl. 24
abundance. Three C·CEIONI·MAXI stamps that have been
previously published with doubtful readings were found in a6) [CLODIAMB]ROSI (Plate 24)
the Slovenian coastal town of Capodistria/Koper.34 There PF: Castrum
are also two fragmented stamps among the finds from Gnirs’ PC: -, BA 4131
excavation of the Roman villa of Valbandond near Fažana.35 Reading: [Q. Clodi Ambr]osi
From Pula Museum we have another two examples on tegula, Lit.: Matijašić 1987, 34
but with the name Q·CEIONIS, derived from the gentilic Desc.: Size 15.6 × 12.5 × 2.9 cm; stamp 4.7 × 2.4 cm.
Ceionio.36
In total we have six examples: one foreign brought from
6. QCLODIAMBROSI the Pula Mausoleum and the others from the Byzantine Cas-
PM: Aquilea and southern Friuli trum on Brijuni Island. All the stamps belong to the same
Dating: first century A.D. variation: no frame, scored letters, no interpunctuation
marks and three groups of nexuses, D^I, A^M and SI.
This is the common form for the marks of Q. Clodius
Ambrosius, the largest private manufacturer of laterite in the
first century A.D., for whom we know of more than a thou-
32
Righini et al. 1993, 64 nos. 19 and 20.
33
Righini et al. 1993, 73 no. 13/1-3. sand stamps scattered around the territories of Aquilea37 and
34
Zaccaria – Zupančić 1993, 140 no. 11.
35
Matijašić 1987, 171 nos. 143-144.
36
Matijašić 1985, 296 no. 8. 37
Cargnelutti 1993, 90-92.

BEZECZKY_Text.indd 153 18.03.19 14:40

 154 EPIGRAPHY

southern Friuli, where his decentralised production centre Reading: A. Faesoni A. f(ili)
was located, with various fornaces around the area.38 His prod- Lit.: Vidimus.
ucts were widely distributed along both sides of the north-
ern Adriatic coast, from Picenum39 and Rimini to Istria and b2) A·FAESONI·A[F] (Plate 25)
southern Dalmatia.40 The importance of his exports in Istria PF: Val Catena
can be clearly seen from the 57 examples documented in PC: BNPAM, Verige s/n
the early collections of the Istria Archaeological Museum,41 Reading: A. Faesoni A. [ f(ili)]
being the fourth most numerous series after the stamps from Lit.: Vidimus.
the Faesoniana, Pansiana and Solonas figlinae.
b3) A·FA[ES]ONI·AF (Plate 25)
7. AFAESONIAF PF: -
PM: Cisalpine Gaul PC: BNPAM, B 91.19
Dating: late first c. B.C. - Julio-Claudian period Reading: A. Fa[es]oni A. f(ili)
Lit.: Vidimus.
a1) A·FAESONI·AF (Plate 24)
PF: - b4) A[·FAES]ONI·AF (Plate 25)
PC: BNPAM, B 91.33 PF: -
Reading: A. Faesoni A. f(ili) PC: BNPAM, B 91.21
Lit.: Vidimus. Reading: A. [Faes]oni A. f(ili)
Lit.: Vidimus.
a2) A[·F]AESONI·AF (Plate 24)
PF: Castrum b5) A[·]FAESONI[·AF] (Plate 25)
PC: -, BA 4139 PF: -
Reading: A. [F]aesoni A. f(ili) PC: BNPAM, s/n
Lit.: Matijašić 1987, 5 Reading: A. Faesoni [A. f(ili)]
Desc.: Size 19.2 × 17.8 × 3.7 cm; stamp 14.5 × 3.3 cm; Lit.: Vidimus.
letter height 2.4 cm.
b6) A[·]FAESO[NI·AF] (Plate 25)
a3) A·FAESONI·A[F] (Plate 24) PF: -
PF: Pula, “via M. Gupca (ex via Castrod)” PC: BNPAM, s/n
PC: BNPAM, A 10649 Reading: A. Faeso[ni A. f(ili)]
Reading: A. Faesoni A. [ f(ili)] Lit.: Vidimus.
Lit.: Vidimus. Matijašić 1987, 43
b7) A[·]FAESO[NI·AF] (Pl. 4)
a4) [A·FAE]SONI·AF (Plate 24) PF: -
PF: Castrum PC: BNPAM, B 91.13
PC: BNPAM, BA 4134 Reading: A. Faeso[ni A. f(ili)]
Reading: [A. Fae]soni A. f(ili) Lit.: Vidimus.
Lit.: Vidimus. Matijašić 1987, 13
b8) A[·]FAES[ONI·AF] (Plate 26)
b1) A·FAESONI·AF (Plate 25) PF: Castrum
PF: - PC: -
PC: BNPAM, B 91.26 Reading: A. Faes[oni A. f(ili)]
Lit.: Vitasović 2007, Sl. 24

b9) A[·]FAES[ONI·AF] (Plate 26)

38
Gomezel 1996, 85 and 90; Zaccaria – Gomezel 2000, 288. PF: Val Catena
39
Marengo 1981.
40
Slapšak 1974. PC: -, A 8392
41
Matijašić 1983, 986. Reading: A. Faes[oni A. f(ili)]

BEZECZKY_Text.indd 154 18.03.19 14:40

 CHAPTER 11 – Inscriptions on laterite finds from Brijuni (Berni Millet) 155

Lit.: Weisshäupl 1899, 82; Matijašić 1987, 3 PC: BNPAM, s/n

Desc.: Thin, excised letters; size 29 × 28 × 2.9 cm; stamp Reading: [A.] Faeso[ni A. f(ili)]
9.2 × 3.6 cm; letter height 2.7 cm Lit.: Vidimus.

b10) A·FAE[SONI·AF] (Plate 26) b18) [A·F]A[ES]ONI[·]AF (Plate 27)

PF: - PF: -
PC: BNPAM, s/n PC: BNPAM, s/n
Reading: A. Fae[soni A. f(ili)] Reading: [A. F]a[es]oni A. f(ili)
Lit.: Vidimus. Lit.: Vidimus.

b11) A·FA[ESONI·AF] (Plate 26) b19) [A·F]AESON[I·AF] (Plate 27)

PF: - PF: -
PC: BNPAM, B 91.2 PC: BNPAM, B 91.12
Reading: A. Fa[esoni A. f(ili)] Reading: [A. F]aeson[i A. f(ili)]
Lit.: Vidimus. Lit.: Vidimus.

b12) A·FA[ESONI·AF] (Plate 26) b20) [A·FA]ESO[NI·AF] (Plate 27)

PF: - PF: Castrum
PC: BNPAM, s/n PC: BNPAM, BA 1500
Reading: A. Fa[esoni A. f(ili)] Reading: [A. Fa]eso[ni A. f(ili)]
Lit.: Vidimus. Lit.: Vidimus.

b13) [A·]FAESONI·AF (Plate 26) b21) [A·FA]ESO[NI·AF] (Plate 27)

PF: - PF: Castrum
PC: BNPAM, s/n PC: BNPAM, BA 1319
Reading: [A.] Faesoni A. f(ili) Reading: [A. Fa]eso[ni A. f(ili)]
Lit.: Vidimus. Lit.: Vidimus.

b14) [A·]FAESONI·AF (Plate 26) b22) [A·F]AESON[I·AF] (Plate 27)

PF: - PF: -
PC: BNPAM, B 91.25 PC: BNPAM, B 91.34
Reading: [A.] Faesoni A. f(ili) Reading: [A. F]aeson[i A. f(ili)]
Lit.: Vidimus. Lit.: Vidimus.

b15) [A·F]AESONI·AF (Plate 26) b23) [A·FA]ESO[NI·AF] (Plate 27)

PF: Castrum PF: Val Madona, Basilica
PC: BNPAM, BA 2952 PC: -, A 8606
Reading: [A. F]aesoni A. f(ili) Reading: [A. Fa]eso[ni A. f(ili)]
Lit.: Vidimus. Matijašić 1987, 6 Lit.: Gnirs 1908, Fig. g; Matijašić 1987, 4
Desc.: Size 12.3 × 12 × 3.3 cm; stamp 7.3 × 3.5 cm.; let-
b16) [A·]FAESO[NI·AF] (Plate 26) ter height 2.8 cm. The stamp was found in the Basilica
PF: - of Val Madona according to Gnirs; Matijašić errone-
PC: BNPAM, s/n ously attributes it to Val Catena.
Reading: [A.] Faeso[ni A. f(ili)]
Lit.: Vidimus. b24) [A·FA]ESO[NI·AF] (Plate 27)
PF: Castrum
b17) [A·]FAESO[NI·AF] (Plate 27) PC: -, BA 1719
PF: - Reading: [A. Fa]eso[ni A. f(ili)]

BEZECZKY_Text.indd 155 18.03.19 14:40

 156 EPIGRAPHY

Lit.: Matijašić 1987, 7 b32) [A·FAES]ONI[·AF] (Plate 28)

Desc.: Size 15 × 12 × 2.9 cm; stamp length 7 cm; letter PF: -
height 2.5 cm. PC: BNPAM, no number
Reading: [A. Faes]oni [A. f(ili)]
b25) [A·FA]ESO[NI·AF] (Plate 27) Lit.: Vidimus.
PF: Castrum
PC: -, BA 1900 b33) [A·FAES]ONI·A[F] (Plate 28)
Reading: [A. Fa]eso[ni A. f(ili)] PF: Castrum
Lit.: Matijašić 1987, 8 PC: -
Desc.: Size 10.5 × 10 × 3; stamp length 10.3 cm; letter Reading: [A. Faes]oni A. [ f(ili)]
height 2.7 cm. Lit.: Vitasović 2007, Sl. 24

b26) [A·FAE]SONI·AF (Plate 28) b34) [A·FAESO]NI·AF (Plate 28)

PF: Castrum PF: Castrum
PC: BNPAM, BA 1905 PC: BNPAM, BA 1897
Reading: [A. Fae]soni A. f(ili) Reading: [A. Faeso]ni A. f(ili)
Lit.: Vidimus. Matijašić 1987, 12 Lit.: Vidimus. Matijašić 1987, 10

b27) [A·FAE]SONI·A[F] (Plate 28) b35) [A·FAESO]NI·AF (Plate 28)

PF: Castrum PF: -
PC: BNPAM, BA 1899 PC: BNPAM, no number
Reading: [A. Fae]soni A. [ f(ili)] Reading: [A. Faeso]ni A. f(ili)
Lit.: Vidimus. Matijašić 1987, 9 Lit.: Vidimus.

b28) [A·FAE]SONI[·]A[F] (Plate 28) b36) [A·FAESO]NI·AF (Plate 28)

PF: Val Catena PF: -
PC: BNPAM, Verige B-VV 26 PC: BNPAM, no number
Reading: [A. Fae]soni A. [ f(ili)] Reading: [A. Faeso]ni A. f(ili)
Lit.: Vidimus. Vitasović 2004, Sl. 4 Lit.: Vidimus.

b29) [A·FAE]SON[I·AF] (Plate 28) b37) [A·FAESON]I[·]AF (Plate 28)

PF: - PF: Castrum
PC: BNPAM, s/n PC: -
Reading: [A. Fae]son[i A. f(ili)] Reading: [A. Faeso]ni A. f(ili)
Lit.: Vidimus. Lit.: Vitasović 2007, Sl. 24

b30) [A·FAES]ONI·AF (Plate 28) b38) [A·FAESON]I·AF (Plate 29)

PF: - PF: -
PC: BNPAM, s/n PC: BNPAM, B 91.36
Reading: [A. Faes]oni A. f(ili) Reading: [A. Faeson]i A. f(ili)
Lit.: Vidimus. Lit.: Vidimus.

b31) [A·FAES]ONI[·]AF (Plate 28) b39) [A·FAESON]I·AF (Plate 29)

PF: - PF: Castrum
PC: BNPAM, B 91.38 PC: BNPAM, BA 4129
Reading: [A. Faes]oni A. f(ili) Reading: [A. Faeson]i A. f(ili)
Lit.: Vidimus. Lit.: Vidimus. Matijašić 1987, 11

BEZECZKY_Text.indd 156 18.03.19 14:40

 CHAPTER 11 – Inscriptions on laterite finds from Brijuni (Berni Millet) 157

b40) [A·FAESONI·]AF (Plate 29) Dalmatia. As far as Istria is concerned, we only have to recall
PF: - the 123 examples from the A·FAESONIA·A·F group depos-
PC: BNPAM, B 91.31 ited in the Pula Archaeological Museum, the second most
Reading: [A. Faesoni] A. f(ili) numerous series after the 201 Pansiana stamps.45 Its vast pro-
Lit.: Vidimus. duction capacity is dated to between the late first century
B.C. and the Julio-Claudian period.
The forty-four A·FAESONI·AF stamps constitute the
8. CFLAVI, CFLAV
largest epigraphic assemblage of all the stamps on laterite
documented to date on Brijuni Island. From the form of the PM: -
texts we can distinguish at least two variations with the same Dating: second half first century B.C.?
lettering. The first variation (a), with four examples, has the
most full-bodied letters, triangular interpunctuation marks a1) C·FLAVI (Plate 29)
and two sets of nexuses, A^E and A^F. The second variation PF: Val Catena, “Westfassade des Terrassenhauses”
(b), with forty examples, is characterised by its thin strokes, PC: -
very small, rounded punctuation marks that are not always Reading: C. Flavi
easy to make out, and a single A^F nexus finalising the in-
Lit.: Gnirs 1908, Fig. C.
scription text. Although for the majority of the pieces there
Desc.: Stamp height 0.2 cm.
is no reference to the place of the find, those that are con-
firmed lead us to believe in a common origin shared between
the archaeological sites of the Castrum and the Roman villa a2) C·FLAVI
of Val Catena. PF: Val Catena, “Zisterne im Peristyle des Terrassen-
All the stamps from Brijuni are impressed on fragments hauses”
of tegulae and use the same formula, A·FAESONI·AF, the PC: -
name of the first owner of one of the most important later- Reading: C. Flavi
ite factories in the eastern part of northern Italy: (ex figlinis) Lit.: Gnirs 1908, 177 (cited).
A(uli) Faesoni A(uli) f(ili). We currently know of three series
written in different ways: FAESONIA, A·FAESONIA·A·F b1) C·FLAVI (Plate 29)
and C·TVLLI·ATISIANI·F/FAESONIA.42 In the second PF: -
series, Righini distinguishes four variations with variable siz- PC: BNPAM, B 91 (no number)
es and diverse sets of nexuses; all the examples from Brijuni
Reading: C. Flavi
can be attributed to his 2d group.
Lit.: Vidimus.
In the traditional academic literature this pottery is
known as Faesonia, but the correct denomination should
be Faesoniana, in keeping with the general derivative form b2) C·F[LAVI] (Plate 29)
of other figlinae names in Cisalpine Gaul, such as Pansiana, PF: -
Cinniana, etc.43 The headquarters of the figlina was between PC: BNPAM, B 91.30
the localities of Rimini and Pesaro (Emilia-Romagna), co- Reading: C. F[lavi]
existing with other major potteries in the region, including Lit.: Vidimus.
Cinniana and Solanas. Based on the number of known ex-
amples and the dissemination of their stamps, Righini44 clas- b3) C·FLAVI (Plate 29)
sified these three potteries in Cisalpine Gaul as medium-size PF: -
figlinae, only smaller than the large Pansiana operation un- PC: BNPAM, B 91.10
der Imperial ownership (cf. no. 12). Reading: C. Flavi
The laterite stamps of the figlina Faesoniana also had a
Lit.: Vidimus.
wide commercial distribution on both coasts of the central
northern Adriatic. Outside Cisalpine Gaul they are found
in large quantities in the areas of Pesaro, Picenum, Istria and b4) [C·F]LAVI + (ante cocturam graffito) (Plate 29)
PF: -
PC: BNPAM, B 91.14
42
Righini 1998, 38-39.
43
Buchi 1987, 152; Righini 1998, 39.
44
Righini 1998, 34. 45
Matijašić 1985, 297.

BEZECZKY_Text.indd 157 18.03.19 14:40

 158 EPIGRAPHY
Reading: C. Flavi
Lit.: Vidimus.
Desc.: To the right of the stamp an ante cocturam graffito
was noted with three strokes giving the appearance of
a cross.
c) C[·]FLAV (Plate 29)
PF: -
PC: BNPAM, B 91.20
Reading: C. Flav(i)
Lit.: Vidimus.
In total six examples were counted with three different
dies for the letterings C·FLAVI and C·FLAV. Variation (a)
is the most regular of all with well squared letters. Variation
(b) was conceived with exactly the same composition, but
with uneven strokes and letter height along the line of writ-
ing. On variation (c) the surface is worn, although it is still
possible to see how the letter L is joined to the foot of the
A, with the latter forming an A^V nexus with the V. The
person referred to in the stamp, C. Flavius, is recorded in
the duo nomina formula, with the name delimited by a small
rounded interpunctuation mark that is clearly visible in the
first two variations.
We only know the exact origin of the examples of the first
variation (a), thanks to Gnirs’ early excavations in the domus
of the Roman villa of Val Catena. The Brijuni Island stamps
are not the only evidence of C·FLAVI documented on tegula
in Istria. Matijašić’s catalogue46 includes a complete exam-
ple of variation (b) discovered in 1965 on the “Via Titograd”
and linked to a suburban Roman villa in the town of Pula.
We do not know where or exactly when this laterite
industry was active. Zaccaria47 believes it could be the
same person who is recorded with the C·FLAVI mark on the
lips of two Dressel 6B amphorae from Padua and Monse-
lice.48 If this connection is confirmed, it would not be the
first instance of the same names appearing on amphorae and
northern Adriatic laterite stamps.49 In our catalogue we list
various confirmed cases of Istrian origin: IMP·AVG·GER
(no. 2), LAEC· (no. 11) and SISENNAE (no. 15); although
the supposed Istrian origin of this other mark is not com-
pletely certain.
The origin of the C·FLAVI amphora stamps listed in
CIL, V. leaves open the possibility that this hypothetical
46
Matijašić 1987, no. 38.
47
Zaccaria 1989, 476.
48
CIL, V. 8112.39a-b.
49
Zaccaria 1989, 476-477.
BEZECZKY_Text.indd 158
mixed industrial pottery production was not Istrian, but
perhaps from the Po Valley, in the southwest of the Veneto
region. It must be taken into account that olive oil amphorae
were also manufactured and stamped on the eastern Padano
plain from the mid first century B.C. and during the first
half of the first century A.D.50 One of the most widely dis-
tributed amphora series in the Po Valley, between Legnano
and Aquilea, pertains to one Flavius Fontanus.51 His stamps
date to around the turn of the era and adopt the forms
FLAV·FONTAN and FONTAN on the lips of Dressel 6B
amphorae.
However, we also have to bear in mind that other mem-
bers of the gens Flavia are recorded on laterite with different
geographical locations to the finds in Italian territory. For
example, we have the case of the C·FLAVIF mark found in
Alvisopoli (Fossalta di Portogruaro) in the southeast of Ven-
ice province.52 We should also take into account the stamps
Q·FLA[---], from San Giovanni in Galilea in the municipal-
ity of Borghi, and L·FLAV/PHILIRI, from Forlì, both in
the region of Emilia-Romagna.53
As far as the possible dating of the C. Flavius stamps on
laterite is concerned, the onomastic formula limited by the
praenomen and nomen, plus the simple, rather careless style
of the lettering, could constitute two valid epigraphic char-
acteristics for suggesting a dating during the second half of
the first century B.C.
9. LFVLLONI
PM: Friuli or Istria
Dating: first century B.C. - first century A.D.
a1) L[·]FVLLO[NI] (Plate 29)
PF: -
PC: BNPAM, B 91.5
Reading: L. Fullo[ni]
Lit.: Vidimus.
a2) [L·F]VLLONI (Plate 29)
PF: Val Catena, “Nordgestade”
PC: -, A 15961
Reading: [L. F]ulloni
Lit.: Gnirs 1908, Fig. e; Matijašić 1987, 2
Desc.: Size 11.5 × 15 × 2.8 cm; stamp 6 × 2.2 cm; letter
height 1.7 cm.
50
Cipriano – Mazzocchin 2012.
51
Cipriano – Mazzocchin 2002.
52
G.A.V.O. 1993, 208.
53
Righini et al. 1993, 69.
18.03.19 14:40
 CHAPTER 11 – Inscriptions on laterite finds from Brijuni (Berni Millet) 159

a3) [L·F]VLLONI (Plate 29) nius Iucundus and Fullonia C. f. Paulla. In fact, from Prati di
PF: - Aiello in the Friuli region, to the northeast of Aquilea, five
PC: BNPAM, B 91.8 examples of L·FVLLONI on reddish-brown tegulae have
Reading: [L. F]ulloni been published by A. Furlan,56 who attributes them to a
Lit.: Vidimus. “quasi certamente istriana” production. However, C. Gome-
zel57 suggests these finds from Aquilea should be linked to
a4) [L·F]VLLO[NI] (Plate 29) the manufacturing zone of the Iulia Concordia ager.58
PF: Castrum In Istria, the gentilic Fullonius is only found on tegulae
PC: BNPAM, BA 1895 stamps. This mark is fairly widely distributed around the
Reading: [L. F]ullo[ni] Istrian Peninsula. The examples from Brijuni Island repre-
Lit.: Vidimus. Matijašić 1987, 28 sent the largest number of finds to date. The early collections
in the Pula Archaeological Museum have five examples.59
a5) [L·FVLL]ONI (Plate 29) Others were found in the localities of Nesazio and Betica.60
PF: - The production of the L·FVLLONI stamps is dated
PC: BNPAM, B 91 (no number) between the centuries before and after the turn of the era,
Reading: [L. Full]oni purely on the basis of epigraphic criteria, mainly the ono-
Lit.: Vidimus. mastic formula of the duo nomina.61

a6) [L·FVLL]ONI (Plate 29) 10. CIVLIAFRICANI

PF: Castrum PM: Aquilea
PC: BNPAM, BA 4128 Dating: second half first century A.D.
Reading: [L. Full]oni
Lit.: Vidimus. Matijašić 1987, 29 a) C·IVLI·AFRIC[ANI] (Plate 29)
PF: Castrum
a7) [L·FVLL]ONI (Plate 29) PC: BNPAM, BA 1893
PF: - Reading: C. Iuli Afric[ani]
PC: BNPAM, no number Lit.: Vidimus. Matijašić 1987, 37
Reading: [L. Full]oni
Lit.: Vidimus. A single incomplete stamp preserved on a tegula frag-
ment from the Brijuni Castrum, published by Matijašić in
We have seven incomplete examples of L·FVLLONI on 1987. The complete text of the C·IVLI·AFRICANI varia-
tegulae fragments. One of the finds is definitely from Val tion is made with deeply scored, neat letters, two triangular
Catena and two are from the Castrum and all were impressed punctuation marks and multiple sets of nexuses: V^L, A^F,
with the same die. The letters are spread out without nexuses R^I, A^N^I.
and use a small circular interpunctuation mark to separate The origin of the laterite production attributed to C. Iulius
the duo nomina of the person recorded.54 Africanus is in the territory of Aquilea with a chronologi-
We currently have very little information about the lat- cal horizon of the second half of the first century A.D. and
erite production of L. Fullonius and that we do have is sub- linked to the activity of the figlina Avitiana.62 Its products
ject to controversy. We do not know with absolute certainty were widely distributed in the territory of Tergeste,63 eastern
where and when this building material industry was active,
although an Istrian or Friulian origin has been hypothesised.
A brief prosopographic study of this family from central
Italy is given by Tassaux,55 who notes their appearance in 56
Furlan 1993, 203 no. 10.
the lapidary epigraphy of Aquilea with the names L. Fullo- 57
Gomezel 1996, 82.
58
Gomezel 1996, 82.
59
CIL, V. 8110.87b; Matijašić 1985, 296 no. 17.
60
Matijašić 1987, 168 nos. 85, 169 no. 116.
54
For a complete example of this variation see Furlan 1993, 201 61
Gomezel 1996, 36.
Fig. 10. 62
Gomezel 1996, 76 and 83.
55
Tassaux 1990, 107. 63
Zaccaria – Zupančić 1993, 144.

BEZECZKY_Text.indd 159 18.03.19 14:40

 160 EPIGRAPHY

Cisalpine Gaul around the Rimini area64 and Istria.65 Accord- b3) LAEC· (Plate 30)
ing to Baldacci,66 the same type of stamp is found impressed PF: Val Catena, “Westflügel Villenanlage”
on amphorae of undefined form in Aquilea and Este, which PC: -, A 6310
is why Zaccaria67 includes it in his prosopographic list of Reading: Laec(ani)
possible manufacturers with mixed, diversified industries Lit.: Gnirs 1904, no. 3; Gnirs 1910, 85, Fig. 8;
producing industrial-type pottery. Matijašić 1987, 1
It has been hypothesised that the person named in these Desc.: Size 12.5 × 16 × 3 cm; stamp 7 × 3.3 cm; nexus
stamps – the owner of the figlina – could have been the A^E
Roman knight and famous orator of Gallic origin, Iulius
Africanus,68 who lived at the time of Nero and was written b4) LAE[C·] (Plate 30)
about by Quintilianus.69 This is an interesting and much PF: Castrum
debated hypothesis70 that remains unresolved due to a lack PC: -, BA 4137
of more solid arguments. Reading: Lae[c(ani)]
Lit.: Matijašić 1987, 23
11. CLAECANIPF, LAEC Desc.: Size 17 × 11.4 × 3.2 cm; stamp 6 × 3.1 cm; letter
PM: Fažana (Istria) height 2.6 cm.
Dating: mid-first century B.C. - 14 A.D.
b5) LAE[C·] (Plate 30)
a) C·LAECANI·P·F (Plate 30) PF: -
PF: Castrum PC: BNPAM, B 91.29
PC: -, BA 4132 Reading: Lae[c(ani)]
Reading: C. Laecani P. f(ili) Lit.: Vidimus.
Lit.: Matijašić 1987, 25
Desc.: Semicircular stamp. Size 16 × 10.2 × 3 cm; stamp b6) [L]AEC· (Plate 30)
8.7 × 4.3 cm; letter height 1.4 cm. PF: Castrum
PC: -, BA 1679
b1) LAEC· (Plate 30) Reading: [L]aec(ani)
PF: Castrum Lit.: Matijašić 1987, 24
PC: -, BA 4140 Desc.: Size 12.5 × 6 × 2.9 cm; stamp 5.3 × 3.1 cm; letter
Reading: Laec(ani) height 2.6 cm.
Lit.: Matijašić 1987, 20
Desc.: Size 12 × 10.2 × 3.2; stamp 6.3 × 3.1 cm; letter b7) [L]AEC[·] (Plate 30)
height 2.6 cm. Rounded punctuation marks. PF: Castrum
PC: -
b2) LAEC· (Plate 30) Reading: [L]aec(ani)
PF: Castrum Lit.: Vitasović 2007, Sl. 24
PC: -, BA 3357
Reading: Laec(ani) b8) [LA]EC· (Plate 30)
Lit.: Matijašić 1987, 21 PF: -
Desc.: Size 12 × 20.5 × 3.1; stamp 6.7 × 3.4 cm; letter PC: BNPAM, no number
height 2.6 cm. Reading: [La]ec(ani)
Lit.: Vidimus.

64
Righini 1998, 65.
c1) LAEC· (Plate 30)
65
Matijašić 1985, 296 no. 19; Matijašić 1987, nos. 60, 138. PF: -
66
Baldacci 1967-68, no. 71. PC: BNPAM, B 91.27
67
Zaccaria 1989, 476.
Reading: Laec(ani)
68
PIR2 I 120.
69
Di Stefano Manzella 1996. Lit.: Vidimus.
70
Zaccaria 1989, 480. Desc.: Rhomboid punctuation marks.

BEZECZKY_Text.indd 160 18.03.19 14:40

 CHAPTER 11 – Inscriptions on laterite finds from Brijuni (Berni Millet) 161

c2) LAEC· (Plate 30) This family’s industrial production in Fažana is well-
PF: - known from the stamps on Dressel 6B amphorae. They
PC: BNPAM, B 91.4 record several generations of Laecanii through an extensive
Reading: Laec(ani) and complex repertory of variations. Bezeczky’s epigraphic
Lit.: Vidimus. catalogue published in 1998 reveals the strength of this busi-
ness in the manufacture of olive oil amphorae, as well as the
c3) LAEC[·] (Plate 30) commercial success of these Istrian products, which were
PF: - very widely distributed in the civil markets of Italy and the
PC: BNPAM, B 91.32 militarised territories of Noricum, Reatia and Pannonia.
Reading: Laec(ani) The Laecanii began to manufacture pottery in Istria in the
Lit.: Vidimus. middle to first century B.C. and continued until the end of
the Julio-Claudian dynasty, when the potteries were taken
c4) LAE[C·] (Plate 30) over by the emperors with the arrival of the Flavian dynas-
PF: Castrum ty.74 It is obvious that the Laecanii’s laterite production has
PC: BNPAM, BA 4136 to be placed during this long historical horizon, between the
Reading: Lae[c(ani)] second half of the first century B.C. and the end of the Julio-
Lit.: Vidimus. Matijašić 1987, 22 Claudian dynasty. The greatest difficulty we have with this
type of epigraphic material is being able to distinguish cor-
d) LAEC (Plate 31) rectly between the different proprietors referred to on the
PF: - stamps and to organise them accurately into the respective
PC: BNPAM, B 91.18 production phases of their figlinae.
Reading: Laec(ani) The interesting C·LAECANI·P·F semicircular mark from
Lit.: Vidimus. variation (a) reads C(aii) Laecan(i) P(ubli) f(ilii). Tassaux75
Desc.: Without punctuation marks next to the letter C. believes this person to be the earliest known to date in his
lineage in Istria. He suggests he could be the founding colonus
There are fourteen examples of tegula stamps from the gens of the Pula colonia from around 46-45 B.C., whereas on the
Laecania. The pictures in our catalogue include at least four other LAEC series we see his son or grandson represented.
variations, one with the longer text C·LAECANI·P·F and Bezeczky76 interprets the C·LAECANI·P·F mark as one of
three with the abbreviated gentilic LAEC, which are quite the first productions of the figlina in a phase of activity prior to
similar but distinguishable, above all from the different ways the turn of the era and suggests dating it to between the mid-
of applying the interpunctuation mark that closes the inscrip- first century B.C. and the year 10 B.C. On the other hand,
tion text. To this list of Laecanii stamps on tegulae we can add he dates the LAEC stamps on tegula to the Augustan period
the C·LAEK·BASS variation preserved in the museum of the between 10 B.C. and 14 A.D., after it had become fashion-
Italian town of Cividale del Friuli (Udine), near the border able to stamp amphorae with the LAEKANI form (with the
with Slovenia, which was published in 1888 by Gregorutti widespread use of the letter K) through an extensive, complex
under number 108 in his epigraphic catalogue.71 and ingenious play of nexuses and abbreviations.
The material evidence in this catalogue originally came One or other of the Laecanii’s stamps on tegula are
from the dwellings owned by this illustrious family on Brijuni known outside their productive and patrimonial area. There
Island,72 including the luxurious seaside villa of Val Catena are a few pieces attesting the LAEC series in Pula,77 one of
and the villa of Val Madona that was subsequently absorbed which came from Gnirs’ early excavations in the Roman
by the walled enclosure of the Byzantine castrum. The Lae- amphitheatre.78
canii’s potteries were located opposite Brijuni Island73 in the
nearby coastal town of Fažana on the southwest of the Istrian
Peninsula.

74
Bezeczky 1998a, 14-28.
75
Tassaux 1998, 83.
71
Bezeczky 1998a, no. 51. 76
Bezeczky 1998a.
72
Tassaux 1982; Tassaux 1998. 77
CIL, V. 8110.101.
73
See Bezeczky in this volume. 78
Matijašić 1987, no. 51.

BEZECZKY_Text.indd 161 18.03.19 14:40

 162 EPIGRAPHY

12. PANSAEVIBI, PANSIANA, TIPANSIANA, a8) PANSAE·VI[BI] (Plate 32)

CCAESARPANS, NERCLAVDPANSIAN PF: Castrum
PM: Cisalpine Gaul PC: BNPAM, BA 1896
Dating: mid-first century B.C. - Tiberius, Caligula, Nero Reading: Pansae Vi[bi]
Lit.: Vidimus. Matijašić 1987, 15
a1) [PAN]SAE·VIBI (Plate 31)
PF: - a9) PANS[AE·VIBI] (Plate 32)
PC: BNPAM, no number PF: Castrum
Reading: [Pa]nsae Vibi PC: BNPAM, BA 1894
Lit.: Vidimus. Reading: Pans[ae Vibi]
Lit.: Vidimus. Matijašić 1987, 16
a2) [PANS]AE[·]VIBI + ((ancora)) (Plate 31)
PF: - a10) PANS[AE·VIBI] (Plate 32)
PC: BNPAM, no number PF: Castrum
Reading: [Pans]ae Vibi PC: BNPAM, BA 1678
Lit.: Vidimus. Reading: Pans[ae Vibi]
Lit.: Vidimus. Matijašić 1987, 17
a3) PANSAE·VIB[I] + ((corona)) (Plate 31)
PF: Castrum a11) PAN[SAE·VIBI] (Plate 32)
PC: BNPAM, BA 2125 PF: Castrum
Reading: Pansae Vib[i] PC: BNPAM, BA 2195
Lit.: Vidimus. Matijašić 1987, 19 Reading: Pan[sae Vibi]
Lit.: Vidimus. Matijašić 1987, 18
a4) [PANS]AE[·]VIBI + ((flos)) (Plate 31)
PF: Castrum a12) PAN[SAE·VIBI ?] (Plate 32)
PC: - PF: -
Reading: [Pansa]e Vibi PC: BNPAM, no number
Lit.: Vitasović 2007, Sl. 24 Reading: Pan[sae Vibi ?]
Lit.: Vidimus.
a5) PANSAE[·VIBI] + ((tridens)) (Plate 31)
PF: Castrum b) PANSIAN[A] (Plate 32)
PC: BNPAM, BA 4133 PF: -
Reading: Pansae [Vibi] PC: BNPAM, no number
Lit.: Vidimus. Matijašić 1987, 14 Reading: Pansian[a]
Lit.: Vidimus.
a6) PANSA[E·VIBI] + ((tridens)) (Plate 31)
PF: Val Catena, “Westfassade des Terrassenhauses, c1) [P]ANSIANA (Plate 32)
Haupteingang” PF: -
PC: - PC: BNPAM, B 91.21
Reading: Pansa[e Vibi] Reading: [P]ansiana
Lit.: Gnirs 1908, Fig. h. Lit.: Vidimus.
Desc.: Stamp height 1.3 cm.
c2) [PANSIAN]A (Plate 32)
a7) PANSA[E·VIBI] + ((tridens)) (Plate 31) PF: -
PF: - PC: BNPAM, no number
PC: BNPAM, B 91.7 Reading: [Pansian]a
Reading: Pansa[e Vibi] Lit.: Vidimus.
Lit.: Vidimus.

BEZECZKY_Text.indd 162 18.03.19 14:40

 CHAPTER 11 – Inscriptions on laterite finds from Brijuni (Berni Millet) 163

d1) [PANSIANA] ((lituus)) (Plate 32) Lit.: Vitasović 2007, Sl. 24

PF: Castrum Desc.: See the complete form in Matijašić 1983, no.
PC: BNPAM, BA 1906 43.
Reading: [Pansiana]
Lit.: Vidimus. i) [---? PAN]SIAN[A] (Plate 33)
Desc.: Only the decorative motif of the Roman lituus PF: Monte Collisi
that embellished the far end of the text PANSIANA PC: BNPAM, Colci B-C 22
has been preserved.79 Reading: [---? Pan]sian[a]
Lit.: Vidimus.
d2) [PA]NSIAN[A ((lituus))] ? (Plate 32)
PF: Pula, Mausoleum The twenty-two stamps from the figlina Pansiana depos-
PC: BNPAM, A 10630 ited in the Brijuni Museum are organised into nine variations
Reading: [Pa]nsian[a] and five large groups that represent various successive histor-
Lit.: Vidimus. Matijašić 1987, 41 ical periods in its significant and long-lasting manufacturing
Desc.: Judging from the form of the preserved letters, activity: PANSAE·VIBI, PANSIANA, TI·PANSIANA,
this may correspond to the PANSIANA variation C·CAESAR·PANS, NERCLAVDPANSIAN.
((lituus)) (Matijašić 1983, no. 15). The PANSAE·VIBI group is the earliest epigraphic evi-
dence and also the most numerous, with twelve examples,
e) [T]I·PANSI[AN]A (Plate 33) seven of which definitely came from the Castrum and one
PF: - from the domus of the Roman villa of Val Catena. This stamp
PC: BNPAM, no number identifies the founder, C. Vibius Pansa, who the figlina Pansi-
Reading: [T]i(beri) Pansi[an]a ana was named after in the mid-first century B.C. Next to his
Lit.: Vidimus. name on the best preserved tegulae we can see other smaller
marks of symbolic meaning (ancora, corona, flos and tridens),
f ) [TI·PANSIA]NA (Plate 33) which may be figurative representations of some of the busi-
PF: Pula, “via Carducci (ora via M. Balote), rivenimento ness managers. The most common group, written as PAN-
anteriore al 1947” SIANA, is considered to be the late Republican and Augus-
PC: BNPAM, A 10626 tan model (ca. 43 B.C.-14 A.D.).81 Of this second group
Reading: [Ti. Pansia]na we have five examples (one of foreign origin from the Pula
Lit.: Vidimus. Matijašić 1987, 44 Mausoleum) and two decorated with the lituus motif with
Desc.: The end of the stamp is preserved with an inverted the spiral pointing towards the right. The remaining groups
nexus, N^A, preceded by the triangular point of the represent the emperors Tiberius (TI·PANSIANA), Caligula
head of the penultimate letter A.80 (C·CAESAR·PANS) and Nero (NERCLAVDPANSIAN).
The Pansiana was the largest Roman pottery attested in
g) [C·C]AESAR·PANS (Plate 33) the territory of Cisalpine Gaul south of the mouth of the
PF: - River Po.82 Over-fired tegulae rejects bearing the name PAN-
PC: BNPAM, no number SIANA have been found in the Voghiera/Voghenza area
Reading: C. Caesar(is) Pans(iana) in the municipality of Ferrara (Emilia-Romagna).83 It was
Lit.: Vidimus. founded by the historical personage, Caius Vibius Pansa
Caetronianus, a friend and supporter of Caesar, governor of
h) [NERCLA]VDP[ANSIAN] (Plate 33) Gallia Cisalpina in the year 45 B.C. and consul in 43 B.C.,
PF: Castrum the same year he was killed in the Battle of Forum Gallo-
PC: - rum against the forces of Mark Anthony. After his death
Reading: [Ner(onis) Cla]ud(i) P[ansian(a)] the figlina became part of the Imperial patrimony, with

81
Righini 1998, 51.
79
For a complete example see Matijašić 1983, no. 15. 82
Righini et al. 1993, 45-52.
80
For the complete TI·PANSIANA form see Matijašić 1983, no. 28. 83
Pellicioni 2012.

BEZECZKY_Text.indd 163 18.03.19 14:40

 164 EPIGRAPHY
three alternative hypotheses vying to explain the change in
ownership:84 a donation to Octavian following its owner’s
death, confiscation by Anthony or the sale of the property
by Pansa’s heirs. This change in ownership brought with
it a major, decisive transformation in the organisation and
development of these industries. From that moment on, the
dissemination of the stamps is manifested as a “monopoly”
on the shores of the Adriatic arc from Picenum to Dalmatia.
The Pansiana form introduced in the stamp after the
death of its founder continued to be used together with the
names of the emperors Tiberius, Caligula, Claudio, Nero,
Galba and Vespasian. Tegulae were the main product of the
figlina Pansiana until the beginning of the Flavian dynasty.
The global productive chronology covers more than a cen-
tury (ca. 45 B.C.-75/80 A.D.), the most intensive produc-
tive phase being during the Imperial period between Augus-
tus and Nero. After that, brick manufacture became more
important, particularly of the so-called sesquipedales, which
were marked with the names of the emperors from Hadrian
to Alexander Severus, although they were manufactured for
a very different market that was restricted to the territory of
Ravenna.85
The notable abundance of stamps from the figlina Pansi-
ana in Istria and their variations served as a pretext in 1983
for Matijašić to undertake the first chronography of this pot-
tery’s stamps, based on the frequent evidence of it found in
the Adriatic regions. Pansiana stamps are the most abundant
in the early collections of the Pula Museum, with 201 exam-
ples, compared to 123 Faesoniana stamps.86
13. LPETRAVIT, LPETRO
PM: Aquilea
Dating: ca. first century B.C. - first century
A.D. (L·PETRO); first century A.D.
(L·PETR·AVIT)
a) L·PETR[·]AVIT (Plate 33)
PF: Castrum
PC: BNPAM, BA 2426
Reading: L. Petr(oni) Avit(i)
Lit.: Vidimus. Matijašić 1987, 32
b1) L·PETRO (Plate 33)
PF: -
84
Righini 1998, 50-51.
85
Righini 1998, 56-58; Pellicioni 2012.
86
Matijašić 1985, 297.
BEZECZKY_Text.indd 164
PC: BNPAM, B 91.28
Reading: L. Petro(ni)
Lit.: Vidimus.
b2) L[·]PETRO (Plate 33)
PF: Val Catena
PC: BNPAM, Verige B-VV 27
Reading: L. Petro(ni)
Lit.: Vidimus. Vitasović 2004, Sl. 4
We have three stamps made with two different dies and
different lettering. The only example of variation (a) comes
from the Castrum and records in scored letters of regular
quality the tria nomina L·PETR·AVIT separated by small
triangular points. Variation (b) is corroborated by two stamps
of poor quality, with the text in a similar style of letter but
in positive relief and the duo nomina L·PETRO separated by
a small circular interpunctuation mark. Found during the
recent excavations at the Roman villa of Val Catena, it is one
of only two examples of the second variation, L·PETRO,
which was previously unknown in Istria. On the other hand,
there are two known parallels for L·PETR·AVIT in Pula and
another three in Nesazio/Nezakcij.87
The Petronii are known from the lapidary epigraphy of
Aquilea where they held public posts in the local government
from the Republican period.88 Likewise, the laterite stamp
from Aquilea with various members of the family represented
in the series has been known for some time: L·PETRONI·L·F,
C·PETRONI·APRI·>·EPIDIAN, L·PETR·AVIT, PETR·
SEC.89 The series with the longest text is read as C. Petroni
Apri > Epidian(a) and is dated to the first century A.D.,
with the graphic particularity of the mark >, which has been
interpreted as indicating the conductor of the figlina Epidi-
ana.90 Current evidence points to one of the laterite produc-
tion zones of the gens Petronia having been in the present-
day locality of Rivignano in Udine province, the origin of
the L·PETRONI·L·F stamp expressed with the filiation that
has been dated to the late Republican period.91 The incised
mark with the tria nomina L·PETR·AVIT is dated later, in
the first century A.D.92 We have no evidence of any finds
outside Istria of the L·PETRO form from the Brijuni tegulae,
87
Matijašić 1987, nos. 53-54 and 105-107.
88
Strazzulla Rusconi 1984, 45, no. 115.
89
CIL, V. 8110.16; Gregorutti 1888, nos. 143-145; Cargnelutti
1992, 105-106; Cargnelutti 1993, 90-97.
90
Gomezel 1996, 40 and 61; Zaccaria – Gomezel 2000, 298-299.
91
Buiatti 1994, 422; Gomezel 1996, 36 and 39.
92
Gomezel 1996, 40.
18.03.19 14:40
 CHAPTER 11 – Inscriptions on laterite finds from Brijuni (Berni Millet) 165

whose onomastic formula – free of cognomen and filiation – Reading: M. Ser[i]

could serve as a pretext to date it between the first century Lit.: Vidimus.
B.C. and the first century A.D.
A family relationship has been proposed between the gens a5) ((corona)) M·SER[I ((corona))] (Plate 34)
Petronia of Aquilea and the powerful Petronii from Salona PF: -
in Dalmatia, or those of Istria in the Flavian–Trajan period, PC: BNPAM, B 91.6
all with close economic ties sustained by a large patrimonial Reading: M. Ser[i]
legacy on both sides of the Adriatic.93 In addition to the lat- Lit.: Vidimus.
erite industry, an activity that linked them was the diversi-
fied production of amphorae for various uses. Perhaps we a6) ((corona)) M·S[ERI ((corona))] (Plate 34)
should look to the Aquilea area for the origin of the Dressel PF: Castrum
6A and 6B amphorae with the P·PETRONI stamp.94 We PC: BNPAM, BA 1901
also have the M·PETRONI·SEC stamp on the handle of an Reading: M. Se[ri]
indeterminate amphora preserved in the early collections of Lit.: Vidimus. Matijašić 1987, 27
the Parma Museum95 that can be linked to the PETR·SEC Desc.: Half way up behind the letter M we can see the
mark on laterite of CIL, V. 8110.208. triangular point with little relief, followed by the let-
ter S with the relief quite damaged.
14. MSERI
PM: - a7) [((corona)) M·S]ERI[ ((corona))] (Plate 34)
Dating: second half first century B.C.? PF: -
PC: BNPAM, B 91.37
a1) ((corona)) M·SERI ((corona)) (Plate 33) Reading: [M. S]eri
PF: Val Catena, “Nordgestade” Lit.: Vidimus.
PC: -
Reading: M. Seri b1) ((corona)) M[·]SERI ((corona)) (Plate 34)
Lit.: Gnirs 1908, Fig. z PF: Val Catena
Desc.: Stamp height 2 cm. PC: BNPAM, Verige B-VV 24
Reading: M. Seri
a2) ((corona)) M[·]SERI ((corona)) (Plate 33) Lit.: Vidimus. Vitasović 2004, Sl. 4
PF: Val Catena
PC: BNPAM, Verige B-VV 25 b2) ((corona)) M[·]SERI ((corona)) (Plate 34)
Reading: M. Seri PF: Val Catena
Lit.: Vidimus. Vitasović 2004, Sl. 4 PC: BNPAM, Verige, no number
Reading: M. Seri
a3) [((corona))] M·SERI ((corona)) (Plate 33) Lit.: Vidimus. Vitasović 2004, Sl. 4
PF: -
PC: BNPAM, B 91.35 b3) [((corona)) M[·]S]ERI ((corona)) (Plate 34)
Reading: M. Seri PF: Val Catena
Lit.: Vidimus. PC: BNPAM, Verige B-VV 29
Reading: [M. S]eri
a4) [((corona))] M·SERI ((corona)) (Plate 33) Lit.: Vidimus. Vitasović 2004, Sl. 4
PF: -
PC: BNPAM, B 91.16 In total there are ten examples of the M·SERI stamp in
two variations that are very similar in their formal aspect
and size. We can distinguish them from the tiny details in
style, for example the uneven way of inscribing the letter
93
Bandelli 1983, 197-198.
94
Cipriano – Mazzocchin 2012, 242. R with the foot prolonged (a) or shortened (b). The text
95
CIL, XI. 6995.69. always appears with two civic crowns decorating each end.

BEZECZKY_Text.indd 165 18.03.19 14:40

 166 EPIGRAPHY
A triangular point separates the duo nomina of the person
M. Serius.
We only know the precise origin of half the finds: one
from the Castrum and four from the Roman villa of Val
Catena. Matijašić’s catalogue96 contains another Istrian
equivalent of our variation (a); it is partially worn and was
found on a fragment of reddish tegula in the Croatian local-
ity of Carnizza/Krnica in Capodistria.
A curiosity to be taken into account is that the first exam-
ple of M·SERI from Val Catena (a1) was published by Gnirs,
together with another SERI stamp impressed on the bottom
of a fragment of terra sigillata and linked to a post cocturam
graffito, ARTE97. Given the unusual and fortuitous coinci-
dence of an infrequent gentilic in laterite production, we ask
ourselves if there could have been a relationship between the
origins of these objects. The stamps on Italic sigillata derived
from the gentilic Serius are well-known among the Po Valley
productions and widely distributed around the towns on the
Adriatic façade of northeastern Italy.98
Based on the simple onomastic formula of the duo nomina
and the decorative detail of the civic crown embellishing the
text, we propose dating this production of unknown origin
to the second half of the first century B.C.
15. SISENNAE
PM: Loron (Istria)
Dating: Tiberius - Claudius
a) SISENNAE (Plate 34)
PF: -
PC: BNPAM, B 91.3
Reading: Sisennae
Lit.: Vidimus.
We do not know the exact origin of the only SISENNAE
stamp found on Brijuni Island. It is on a fragment of tegula
and is an important new find for the research, as it can be
linked to the SISENNAE stamps on the Dressel 6B ampho-
rae manufactured in Loron (Istria), which have been accu-
rately dated to the reigns of Tiberius and Claudius in the
Austrian locality of Magdalensberg.99 To date, the amphora
stamps have always been found with the same lettering –
96
Matijašić 1987, 172 no. 157.
97
Gnirs 1908, Fig. h.
98
Oxé – Comfort 1968, nos. 1752-1773; Oxé – Comfort 2000,
nos. 1882-1906.
99
Bezeczky 1994, 96-98.
BEZECZKY_Text.indd 166
SISENNAE – but with the text enclosed between decora-
tive motifs – a palma on the left and a corona on the right.
Behind this name we find the historical figure of
T. Sisenna Statilius Taurus, a first century A.D. Roman
senator who went through his cursus honorum during the
empires of Augustus and Tiberius. He was also the grandson
of T. Statilius Taurus, an important general under Augustus
and an ordinary consul in 37 and 26 B.C. Sisenna himself
was named ordinary consul under Tiberius in 16 A.D.100 He
was married to Cornelia, with whom he had two children,
Sisenna Statilius Taurus101 and Statilia Cornelia.
The SISENNAE stamps on olive oil amphorae – and now
for the first time on tegula – provide clear archaeological and
epigraphic evidence of the rich patrimony of this family in
Istria. Sisenna, T. Statilius Taurus’ son, owned various estates
in the so-called ager Parentinus, between the coastal area of
Loron and the inland locality of Motovun/Montona, as well
as further north in Piquentum in the territory of Trieste.102
The Dressel 6B amphorae with the SISENNAE stamps
were manufactured in Loron when that important figlina
was owned by the wealthy senator. It later passed into the
hands of the courtesan Calvia Crispinilla, a friend of Nero
(“magistra libidinum Neronis”, Tacitus, Hist. 1, 73). In Sisen-
na’s time it manufactured a wide range of pottery products
and his name can be found on amphorae, tegulae and terra
sigillata vessels.103 The pottery passed into imperial hands
during the reign of Domitian (vide no. 2) and from then on
the stamps bore the names of the emperors up to the time of
Hadrian.
16. LSTIVSTI
PM: Aquilea
Dating: from the mid-first century A.D.
a) [L·]ST·IVSTI (Plate 34)
PF: Val Catena, “Tempelbezirk”
PC: -
Reading: [L.] St(ati, -atili) Iusti
Lit.: Gnirs 1908, Fig. b.
Desc.: Incised 2.2 cm-high letters.
The complete stamp L·ST·IVSTI is found without a
frame, with elegantly incised, well-delineated characters, two
sets of nexuses, S^T and T^I and the tria nomina divided
100
PIR1 S 613.
101
PIR1 S 614.
102
Tassaux 2005, 142.
103
Tassaux et al. 2001; Maggi – Marion 2007.
18.03.19 14:40
 CHAPTER 11 – Inscriptions on laterite finds from Brijuni (Berni Millet) 167

by circular points.104 Based on the variable size of the letters,
three different dies with the same shape and lettering have
been distinguished.105
Tassaux106 includes the Statii among the families of the
Aquilea municipal aristocracy with laterite manufacturing
industries. It is precisely in the western territory of Aquilea
that the C·STAT, C·STATI·IVSTI, L·STATI, L·ST·IVSTI
stamps are found most frequently, along with other marks
alluding to the figlina Statiana that are dated to the mid-first
century A.D.107 The productive activity of L. Statius Iustus is
dated to the second half of the first century A.D.108 Another
laterite manufacturer from Aquilea linked to this family is
the person on the ST·SILON stamp, interpreted as Statii
Silonis in Gregorutti’s catalogue.109
Very few Istrian parallels have been documented to date.
We have already mentioned the unprecedented example in
Matijašić’s catalogue which came from Lisignano/Ližnjan,
south of Pula, on the southern tip of the peninsula. Another
definite find with an incomplete text [---]IVS[---] is from
the Roman villa of Dragonera, north of Fažana.110
17. TROSI
PM: Friulian territory
Dating: late first century B.C. - early first century A.D.
a1) TROSI (Plate 34)
PF: Val Catena, “Villenanlagen”
PC: -
Lit.: Reading: Trosi
Gnirs 1904, 144 no. 2.
Desc.: Yellowish imbrex.
a2) TROSI (Plate 34)
PF: -
PC: BNPAM, B 91.9
Reading: Trosi
Lit.: Vidimus.
a3) [TR]OSI (Plate 34)
PF: Castrum
PC: BNPAM, BA 1901
Reading: [Tr]osi
Lit.: Vidimus. Matijašić 1987, 31
We have three stamps from the TROSI series, all with the
same variation without frame and with thin-bodied incised
letters. The first example (a), excavated by Gnirs in the
Roman villa of Val Catena, corresponds to an imbrex, while
the others are stamped on fragments of tegulae.
The gens Trosia is widely attested in the lapidary epigra-
phy of Aquilea and Trieste.111 The geographic dispersion of
this family and the distribution of the laterite signed with
the marks TROSI, C·TROSI, M·TROSI, P·TROSI is sig-
nificant along the northern Adriatic coast, demonstrating
a commercial vocation comparable to that of the Barbii
(vide no. 3), who covered the region of Friuli and the Istrian
Peninsula. The P·TROSI series is dated to the first century
B.C. and the others –TROSI, C·TROSI and M·TROSI – to
between the late first century B.C. and the early first century
A.D.112
The Trosii’s business activities appear to correspond to
another decentralised productive phenomenon in the terri-
tory of Friuli. Their stamps are found relatively frequently in
Aquilea; the letters are in relief or incised, with or without
frame, with the forms TROSI, P·TROSI and M·TROSI,
with or without alternation of praenomen.113 In the eastern-
most area of the Veneto, the TROSI series is also considered
among the best-known stamps on laterite.114 Other suggested
possible production areas are the towns of Porpetto, Car-
lino and San Giorgio di Nogaro (Udine province) between
the first century B.C. and the beginning of the first century
A.D.115 Another production site was excavated by M. Stokin
near San Bernardino in the coastal locality of Pirano/Piran
in southwestern Slovenia, formerly part of the territory of
Roman Tergeste.116
As far as distribution in Istria is concerned, in addi-
tion to the TROSI example in the Pula Museum cited by
Matijašić,117 we have to recall the other stamp on reddish
tegula found in the Roman villa of Valbandon.118

111
Tassaux 1990, 105.
104
See the complete Istrian example from Lisignano/Ližnjan in 112
Gomezel 1996, 36 and 46.
Matijašić 1987, 173, no. 161. 113
Cargnelutti 1992, 109; Cargnelutti 1993, 99; Furlan 1993,
105
Cargnelutti 1992, 107; Cargnelutti 1993, 99. 199-200.
106
Tassaux 1990, 106. 114
G.A.V.O. 1993, 208.
107
Gomezel 1996, 36, 40 and 52. 115
Gomezel 1996, 91; Vazzoler 2009.
108
Buiatti 1990-91, 191; Gomezel 1996, 40. 116
Zaccaria – Zupančić 1993, 150 no. 73.
109
Gregorutti 1888, no. 174. 117
Matijašić 1985, 296 no. 39.
110
Starac 2010b, 111, T.III.3. 118
Matijašić 1987, no. 151.

BEZECZKY_Text.indd 167 18.03.19 14:40

 168 EPIGRAPHY

B. EPIGRAPHY ON DOLIA
18. ARIAL[---]
PM: Veneto?
Dating: -

Fig. 11.2 Post cocturam capacity graffiti on the dolia

a) ARIA L[---] (Plate 34)
PF: -
PC: BNPAM, BA 1681
Reading: Aria L. [------]
Lit.: Vidimus.
The stamp is fragmented on the right. We can clearly see
two words separated by a sufficiently wide blank space and
without visible punctuation marks. Of the single remaining
letter of the next word only the upper part of the vertical
shaft is preserved; in our view it appears to correspond to
an L.
The gentilic Ar(r)ius is widely documented in Cisalpine
Gaul, Picenum and Venetia. Various members of this family
are well-represented in the lapidary inscriptions of Aquilea
and were already part of the municipal aristocracy in the
Republican period.119 Their presence in Trieste and north-
eastern Istria is equally notable in the early Roman period in
the first century A.D., when the lapidary inscriptions men-
tion various members of this gens of different genders and
social statuses (a woman, a sevir and two freedmen).120
A stamp, ARIAE (with the nexus A^E), on what is
assumed to be laterite, refers to a woman121 from the Slove-
nian area of Servola/Škedenj near Trieste.122 From the Vene-
tia area there are other laterite stamps representing various
people with the gentilic Arrius, which is why it has been sug-
gested that the production zone of the figlina Arriana was
somewhere between Padua and Aquilea123. From the formal
traits of the text, these stamps can be dated to within the first
half of the first century A.D.
from Val Catena (Gnirs 1908).
A post cocturam graffito preserved on a small fragment of
the upper part of a large dolium. The letters are tall, shallow
and thin-bodied. The principal unit LXIIII is expressed in
amphorae and is almost completely preserved. Taking into
account the value 26.2656 equivalent to the amphora as a
unit of measure, this dolium must have had a capacity of
around 1,681 litres.
In 1908 Gnirs published an interesting set of post coc-
turam graffiti on dolia, three from the cella vinaria of Val
Catena (a, c, d) and a fourth from the cella olearia of Monte
Collisi (b). Note in the following illustration how in all of
them the foot of the letter L is scored with a prolonged hori-
zontal stroke on both sides. In the example from the Brijuni
Museum the base of the letter L is incomplete on the left,
but there is a high probability that it is the same style of let-
ter. Gnirs’ four graffiti give us the following litre capacity
values: LVIII = 1,523 (a); LXI (et) s(emis) = 1,615.33 (b);
LXVII = 1,785 (c); LXV = 1,707.2 (d). Given the homog-
enous system recording these Figs. on the dolia from Val
Catena, our graffito must be complete (LXIIII), unless the
fraction s(emis) of half an amphora (1,615.33 litres) should
also be indicated at the end of the text.

19. (amphorae) LXIIII [--- ?]

PF: Castrum
PC: BNPAM, no number
Reading: (amphorae) LXIIII [--- ?]
Lit.: Vidimus.

119
Tassaux 1990, 82-83.
120
Zaccaria – Zupančić 1993, 163-164.
121
Gomezel 1996, 47.
122
CIL, V. 8110 183; Zaccaria – Zupančić 1993, 139 no. 4.
123
Righini 1998, 35.

BEZECZKY_Text.indd 168 18.03.19 14:40

 CHAPTER 11 – Inscriptions on laterite finds from Brijuni (Berni Millet) 169

C. GENERAL INDEX
No. Var. Stamp Inv. no. Place of the find
01 a [M·ALBI·R]VFI A 10643 Pula, Campidoglio Theatre
02 a IMP·AVG·G[ER] no number –
L·BARBI·L·L·EVPOR[IS] &
03 a A 10628 Pula, Amphitheatre “maggio, 1938”
L·BARBI·L·L·EVPOR[I]
04 a1 Q·G·NICEP B 91.22 –
04 a2 Q·G·NIC[EP] B 91.1 –
04 a3 Q·G·NIC[EP] BA 1903 Castrum
04 a4 [Q·]G·NICEP B 91.11 –
04 a5 [Q·]G·NICEP B 91.15 –
05 a1 C·CEIONI·MAXI – Monte Collisi
05 a2 [C·]CEIONI[·]MAXI BA 4130 Castrum
06 a1 QCLODIAMB[ROSI] BA 9447 Castrum
06 a2 QCLODIAM[BROSI] BA 1904 Castrum
06 a3 QCLODI[AMBROSI] A 10629 Pula, Mausoleum
06 a4 QCLO[DIAMBROSI] BA 4135 Castrum
06 a5 [QC]LODIAMBRO[SI] – Castrum
06 a6 [CLODIAMB]ROSI BA 4131 Castrum
07 a1 A·FAESONI·AF B 91.33 –
07 a2 A[·F]AESONI·AF BA 4139 Castrum
07 a3 A·FAESONI·A[F] A 10649 Pula, “via M. Gupca (ex via CastroPula)”
07 a4 [A·FAE]SONI·AF BA 4134 Castrum
07 b01 A·FAESONI·AF B 91.26 –
07 b02 A·FAESONI·A[F] Verige no number Val Catena
07 b03 A·FA[ES]ONI·AF B 91.19 –
07 b04 A[·FAES]ONI·AF B 91.21 –
07 b05 A[·]FAESONI[·AF] no number –
07 b06 A[·]FAESO[NI·AF] no number –
07 b07 A[·]FAESO[NI·AF] B 91.13 –
07 b08 A[·]FAES[ONI·AF] – Castrum
07 b09 A[·]FAES[ONI·AF] A 8392 Val Catena
07 b10 A·FAE[SONI·AF] no number –
07 b11 A·FA[ESONI·AF] B 91.2 –
07 b12 A·FA[ESONI·AF] no number –
07 b13 [A·]FAESONI·AF no number –
07 b14 [A·]FAESONI·AF B 91.25 –
07 b15 [A·F]AESONI·AF BA 2952 Castrum
07 b16 [A·]FAESO[NI·AF] no number –
07 b17 [A·]FAESO[NI·AF] no number –
07 b18 [A·F]A[ES]ONI[·]AF no number –
07 b19 [A·F]AESON[I·AF] B 91.12 –
07 b20 [A·FA]ESO[NI·AF] BA 1500 Castrum
07 b21 [A·FA]ESO[NI·AF] BA 1319 Castrum
07 b22 [A·F]AESON[I·AF] B 91.34 –
07 b23 [A·FA]ESO[NI·AF] A 8606 Val Madona, Basilica
07 b24 [A·FA]ESO[NI·AF] BA 1719 Castrum

BEZECZKY_Text.indd 169 18.03.19 14:40

 170 EPIGRAPHY

No. Var. Stamp Inv. no. Place of the find

07 b25 [A·FA]ESO[NI·AF] BA 1900 Castrum
07 b26 [A·FAE]SONI·AF BA 1905 Castrum
07 b27 [A·FAE]SONI·A[F] BA 1899 Castrum
07 b28 [A·FAE]SONI[·]A[F] Verige B-VV 26 Val Catena
07 b29 [A·FAE]SON[I·AF] no number –
07 b30 [A·FAES]ONI·AF no number –
07 b31 [A·FAES]ONI[·]AF B 91.38 –
07 b32 [A·FAES]ONI[·AF] no number –
07 b33 [A·FAES]ONI·A[F] – Castrum
07 b34 [A·FAESO]NI·AF BA 1897 Castrum
07 b35 [A·FAESO]NI·AF no number –
07 b36 [A·FAESO]NI·AF no number –
07 b37 [A·FAESON]I[·]AF – Castrum
07 b38 [A·FAESON]I·AF B 91.36 –
07 b39 [A·FAESON]I·AF BA 4129 Castrum
07 b40 [A·FAESONI·]AF B 91.31 –
08 a1 C·FLAVI – Val Catena, “Westfassade des Terrassenhauses”
Val Catena, “Zisterne im Peristyle des Terrassen-
08 a2 C·FLAVI, [no sketch] –
hauses”
08 b1 C·FLAVI B 91 (no number) –
08 b2 C·F[LAVI] B 91.30 –
08 b3 C·FLAVI B 91.10 –
08 b4 [C·F]LAVI + (ante cocturam graffito) B 91.14 –
08 c C[·]FLAV B 91.20 –
09 a1 L[·]FVLLO[NI] B 91.5 –
09 a2 [L·F]VLLONI A 15961 Val Catena, “Nordgestade”
09 a3 [L·F]VLLONI B 91.8 –
09 a4 [L·F]VLLO[NI] BA 1895 Castrum
09 a5 [L·FVLL]ONI no number –
09 a6 [L·FVLL]ONI BA 4128 Castrum
09 a7 [L·FVLL]ONI B 91 (no number) –
10 a C·IVLI·AFRIC[ANI] BA 1893 Castrum
11 a C·LAECANI·P·F BA 4132 Castrum
11 b1 LAEC· BA 4140 Castrum
11 b2 LAEC· BA 3357 Castrum
11 b3 LAEC· A 6310 Val Catena, “Westflügel Villenanlage”
11 b4 LAE[C·] BA 4137 Castrum
11 b5 LAE[C·] B 91.29 –
11 b6 [L]AEC· BA 1679 Castrum
11 b7 [L]AEC[·] – Castrum
11 b8 [LA]EC· no number –
11 c1 LAEC· B 91.27 –
11 c2 LAEC· B 91.4 –
11 c3 LAEC[·] B 91.32 –
11 c4 LAE[C·] BA 4136 Castrum

BEZECZKY_Text.indd 170 18.03.19 14:40

 CHAPTER 11 – Inscriptions on laterite finds from Brijuni (Berni Millet) 171

No. Var. Stamp Inv. no. Place of the find

11 d LAEC B 91.18 –
12 a01 [PAN]SAE·VIBI no number –
12 a02 [PANS]AE[·]VIBI + ((ancora)) no number –
12 a03 PANSAE·VIB[I] + ((corona)) BA 2125 Castrum
12 a04 [PANS]AE[·]VIBI + ((flos)) – Castrum
12 a05 PANSAE[·VIBI] + ((tridens)) BA 4133 Castrum
Val Catena, “Westfassade des Terrassenhauses,
12 a06 PANSA[E·VIBI] + ((tridens)) –
Haupteingang”
12 a07 PANSA[E·VIBI] + ((tridens)) B 91.7 –
12 a08 PANSAE·VI[BI] BA 1896 Castrum
12 a09 PANS[AE·VIBI] BA 1894 Castrum
12 a10 PANS[AE·VIBI] BA 1678 Castrum
12 a11 PAN[SAE·VIBI] BA 2195 Castrum
12 a12 PAN[SAE·VIBI ?] no number –
12 b PANSIAN[A] no number –
12 c1 [P]ANSIANA B 91.21 –
12 c2 [PANSIAN]A no number –
12 d1 [PANSIANA] ((lituus)) BA 1906 Castrum
12 d2 [PA]NSIAN[A ((lituus))] ? A 10630 Pula, Mausoleum
12 e [T]I·PANSI[AN]A no number –
Pula, “via Carducci (ora via M. Balote),
12 f [TI·PANSIA]NA A 10626
rivenimento anteriore al 1947”
12 g [C·C]AESAR·PANS no number –
12 h [NERCLA]VDP[ANSIAN] – Castrum
12 i [---? PAN]SIAN[A] Colci B-C 22 Monte Collisi
13 a L·PETR[·]AVIT BA 2426 Castrum
13 b1 L·PETRO B 91.28 –
13 b2 L[·]PETRO Verige B-VV 27 Val Catena
14 a1 ((corona))M·SERI ((corona)) – Val Catena, “Nordgestade”
14 a2 ((corona))M[·]SERI ((corona)) Verige B-VV 25 Val Catena
14 a3 [((corona))]M·SERI ((corona)) B 91.35 –
14 a4 [((corona))]M·SERI ((corona)) B 91.16 –
14 a5 ((corona))M·SER[I ((corona))] B 91.6 –
14 a6 ((corona))M·S[ERI ((corona))] BA 1901 Castrum
14 a7 [((corona)) M·S]ERI[ ((corona))] B 91.37 –
14 b1 ((corona))M[·]SERI ((corona)) Verige B-VV 24 Val Catena
14 b2 ((corona))M[·]SERI ((corona)) Verige no number Val Catena
14 b3 [((corona))M[·]S]ERI ((corona)) Verige B-VV 29 Val Catena
15 a SISENNAE B 91.3 –
16 a [L·]ST·IVSTI – Val Catena, “Tempelbezirk”
17 a1 TROSI – Val Catena, “Villenanlagen”
17 a2 TROSI B 91.9 –
17 a3 [TR]OSI BA 1901 Castrum
18 a ARIAL[---] BA 1681 Castrum
19 – (amphorae) LXIIII [--- ?] – –

BEZECZKY_Text.indd 171 18.03.19 14:40

BEZECZKY_Text.indd 172 18.03.19 14:40
 PLATE 23

1 2

3 4a1

4a2
4a3

4a4 4a5

5a1 5a2

6a1

[M·ALBI·R]VFI (no. 1), IMP·AVG·G[ER] (no. 2), L·BARBI·L·L·EVPOR[IS] & L·BARBI·L·L·EVPOR[I] (no. 3),
Q·G·NICEP (nos. 4a1, 4a2, 4a3, 4a4, 4a5), C·CEIONI·MAXI (nos. 5a1, 5a2), Q·CLODI·AMBROSI (no. 6a1).

BEZECZKY_Text.indd 23 18.03.19 14:40

 PLATE 24

6a2 6a3

6a4 6a5

6a6 7a1

7a2

7a3

7a4

Q·CLODI·AMBROSI (nos. 6a2, 6a3, 6a4, 6a5, 6a6), A·FAESONI·AF (nos. 7a1, 7a2, 7a3, 7a4).

BEZECZKY_Text.indd 24 18.03.19 14:40

 PLATE 25

7b1

7b2

7b3

7b4

7b5

7b6

A·FAESONI·AF (nos. 7b1, 7b2, 7b3, 7b4, 7b5, 7b6).

BEZECZKY_Text.indd 25 18.03.19 14:40

 PLATE 26

7b7 7b8

7b9 7b10 7b11

7b12 7b13

7b14

7b15

7b16

A·FAESONI·AF (nos. 7b7, 7b8, 7b9, 7b10, 7b11, 7b12, 7b13, 7b14, 7b15, 7b16).

BEZECZKY_Text.indd 26 18.03.19 14:40

 PLATE 27

7b17

7b18

7b19 7b20

7b21 7b22

7b23-Gnirs 1908 7b23-Matijašić 1987

7b24 7b25

A·FAESONI·AF (nos. 7b17, 7b18, 7b19, 7b20, 7b21, 7b22 ,7b23-Gnirs 1908, 7b23-Matijašić 1987, 7b24, 7b25).

BEZECZKY_Text.indd 27 18.03.19 14:40

 PLATE 28

7b26 7b27

7b28 7b29

7b30 7b31

7b32 7b33 7b34

7b35 7b36 7b37

A·FAESONI·AF (nos. 7b26, 7b27, 7b28, 7b29, 7b30, 7b31, 7b32, 7b33, 7b34, 7b35, 7b36, 7b37).

BEZECZKY_Text.indd 28 18.03.19 14:40

 PLATE 29

7b38 7b39 7b40

8a1
8b1 8b2

8b3
8b4

8c 9a1 9a2-Gnirs 1908

9a2-Matijašić 1987 9a3 9a4

9a5 9a6 9a7

10

A·FAESONI·AF (nos. 7b38, 7b39, 7b40), C·FLAVI (nos. 8a1, 8b1, 8b2, 8b3, 8b4, 8c), L·FVLLONI (nos. 9a1, 9a2-Gnirs 1908,
9a2-Matijašić 1987, 9a3, 9a4, 9a5, 9a6, 9a7), C·IVLI·AFRICANI (no. 10).

BEZECZKY_Text.indd 29 18.03.19 14:40

 PLATE 30

11b1

11a

11b2 11b3- Gnirs 1904/1910 11b3-Matijašić 1987

11b4 11b6
11b5

11b7 11b8 11c1

11c2 11c3 11c4

C·LAECANI·P·F (no. 11a), LAEC (nos. 11b1, 11b2, 11b3-Gnirs 1904/1910, 11b3-Matijašić 1987, 11b4, 11b5, 11b6, 11b7,
11b8, 11c1, 11c2, 11c3, 11c4).

BEZECZKY_Text.indd 30 18.03.19 14:40

 PLATE 31

11d 12a1

12a2 12a3

12a4

12a5

12a6
12a7

LAEC (no. 11d), PANSAE·VIBI (nos. 12a1, 12a2, 12a3, 12a4, 12a5, 12a6, 12a7).

BEZECZKY_Text.indd 31 18.03.19 14:40

 PLATE 32

12a8 12a9

12a10 12a11 12a12

12b

12c1 12c2

12d1 12d2

PANSAE·VIBI (nos. 12a8, 12a9, 12a10, 12a11, 12a12), PANSIANA (12b, 12c1, 12c2, 12d1, 12d2).

BEZECZKY_Text.indd 32 18.03.19 14:40

 PLATE 33

12e 12f

12h
12g

12i 13a

13b1 13b2

14a1

14a2

14a3

14a4

TI·PANSIANA (nos. 12e, 12f ), C·CAESAR·PANS (no. 12g), NER·CLAVD·P (no. 12h), PANSIAN (no. 12i),
L·PETR·AVIT (no. 13a), L·PETR (nos. 13b1, 13b2), M·SERI (nos. 14a1, 14a2, 14a3, 14a4).

BEZECZKY_Text.indd 33 18.03.19 14:40

 PLATE 34

14a5 14a6

14a7
14b1

14b2
14b3

16
15

17a1 17a2

17a3

18 19

M·SERI (nos. 14a5, 14a6, 14a7, 14b1, 14b2, 14b3), SISENNAE (no. 15), L·ST·IVSTI (no. 16), TROSI (nos. 17a1, 17a2, 17a3),
dolium stamp ARIA(…) (no. 18), dolium graffito LXIII(…) (no. 19).

BEZECZKY_Text.indd 34 18.03.19 14:40

 MICROPETROGR APHY

BEZECZKY_Text.indd 185 18.03.19 14:40

BEZECZKY_Text.indd 186 18.03.19 14:40

12 MICROPETROGRAPHY OF THE FAŽANA AMPHORAE
György Szakmány – Sándor Józsa
Petrography is the branch of petrology (within geology)
dealing with the description and classification of rocks,
especially by microscopic examination, without neglecting
macroscopic description. Micropetrography in our interpre-
tation here deals only with the polarising microscopic study
of thin sections of rocks and any other materials containing
components or fragments of rocks. Fired pottery, such as,
for example, in our case the amphorae, are made of a mash
of rocks or rock debris, which contains plastic and non-
plastic components. The plastic material usually constitutes
more than 70 % of the whole raw material, which consists of
submicroscopic material and thus it cannot be determined
exactly by polarising microscope. Accordingly, the main
subjects of our micropetrographic study are the non-plastic
components of the Fažana amphorae.
The amphorae of Castrum villa make up only a very small
part of all known amphorae manufactured in Fažana. In spite
of this, on the basis of our petrographic investigations, we
can say that Castrum amphorae fit well with the main line of
all Fažana amphorae.1 It is obvious that Castrum amphorae
carry enough characteristics to show clearly what were the
main raw materials for not only the amphorae of Castrum
villa, but also for all other amphorae manufactured in Fažana
in the period under study. In this work, we describe the ear-
lier and the new results of Fažana amphora research. Exten-
sive petrographic properties of Castrum amphorae are listed
in the catalogue. Here we present only those micropetro-
graphic features which are the most useful for determining
the raw material. In the list of different determinable com-
ponents below, we also use some data which are poorly rep-
resented in the Castrum amphorae. Otherwise, all amphora
illustrations were taken from Castrum villa samples.
The main goal of our present study is to obtain as much
knowledge as possible about the raw materials of the ampho-
rae and the production technology using only micropetrog-
raphy, namely: 1) type of the raw materials, 2) ratio of the
1
Józsa et al. 2016.
BEZECZKY_Text.indd 187
187
raw materials in the mixture, 3) possible provenance of the
raw materials.
PREVIOUS WORKS ON RAW MATERIAL DETER-
MINATION
The archaeometric research on the Dressel 6B amphorae
began more than 30 years ago. At the beginning, about 200
thin sections of amphorae were quantitatively and quali-
tatively described. The first classification was based on the
texture and the non-plastic components.2 In addition, a few
samples were measured by XRD for a coarse evaluation of
the firing temperature.3 Later on, detailed micromineralogi-
cal analysis of the amphorae was performed for comparison
with the terra rossa and loess from the Istrian peninsula. At
the same time, 300 new thin sections of amphorae have been
evaluated, and a new classification has been created based on
the previous and the new results.4
We summarize here the main results of the research men-
tioned above. 1) Nine petrographic groups were determined
based on the fabric analysis. 2) The main component of the
amphorae was found to be similar to the loess-containing
terra rossa from Istria. 3) Research showed recent marine
sediments and plankton in the raw material of the ampho-
rae. 4) The heavy mineral composition of the amphorae
samples and the loess-containing terra rossa were found to
be partly different. The origin of these heavy minerals in the
amphorae remains unknown. 5) The firing temperature was
roughly estimated in the range of 750-900 oC.
METHODS AND STRATEGY
Recently newer and increasingly more accurate (and inciden-
tally not more expensive) research methods are available to
2
Bezeczky 1987; Józsa – Szakmány 1987.
3
Weiszburg – Papp 1987.
4
Mange – Bezeczky 2006; Mange – Bezeczky 2007;
Bezeczky – Mange 2009.
18.03.19 14:40
 188 MICROPETROGRAPHY
the scholars who are working on the archaeometry of ceram-
ics, in our case on the determination of the raw materials of
ceramics. These methods give very precise data, but mostly
do not allow direct and comprehensive observation of the
whole sample studied in its original condition.
There are only a few research methods in archaeometry
(petrography, electron microscopy), which allow a direct
view of the different components which were used to com-
pose the raw material for amphora production as a whole.
In this study, we used only petrographic analysis of thin
sections to show that this method is sufficient by itself for
the exact observation, determination, comparison and corre-
lation of most different geological materials. We established
a new petrographic research strategy using the advantages
mentioned above and based on the fact that petrography is
qualified to investigate a huge number of samples compara-
tively cheaply.
In the course of a detailed study of hundreds of thin sec-
tions of Fažana amphorae, we were able to find quite a few
samples in which larger whole rock fragments of different
raw materials could be observed as inclusions. During the
detailed petrographic study, these rock fragments in ampho-
rae were compared with the different field rock samples col-
lected from the possible source territories. Detailed whole
rock petrography gave the best results for raw material iden-
tification. The raw materials of amphorae without whole
rock inclusions could only be determined by detailed exami-
nation of the individual non-plastic components. Firstly, we
attempted to find genetic connections between the indi-
vidual components and whole rock fragments in amphorae
and rock samples from the field. After that, we were able to
obtain information about the raw materials, even from those
individual grains which are less informative alone than the
whole rock fragments, but appear in almost all amphora
samples.
With this method as a first step, we identified the main
components and most likely localities of the raw materials
of Fažana amphorae. As a second step, we determined the
different main components of the raw material in almost
every single amphora fragment which had an available thin
section.
To successfully implement our petrographic strategy
we also applied a new approach to our long-term research,
slightly different from the protocol used by previous
researchers.
1. After a macroscopic grouping, a brief description of
texture and observation of the main non-plastic components
BEZECZKY_Text.indd 188
in the thin sections of representative amphorae were made
by polarising microscope. At this stage of the research only a
preliminary grouping is required.
2. Detailed study of the geological literature (articles, field
guides, geological maps etc.) of the wider area surrounding
the amphora workshop was conducted.
3. For extensive recognition of different geological for-
mations in the field we searched for outcrops of typical rock
types of different formations, and collected rock samples,
primarily for making thin sections. During the sampling, we
preferred materials which seemed to be suitable for pottery
production. Later on, we tried to find deposits of possible
raw materials which were suitable for extraction in a larger
quantity and from where the raw material would be easy to
ship.
4. A detailed petrographic microscopic study of all possi-
ble raw materials from the potential source areas (in our case
in the Istrian peninsula and surroundings) was completed.
5. A detailed comparative petrographic study of all sam-
ples of amphorae and raw materials was undertaken. In this
stage of the study a grouping can be made based on the pro-
portions of the different raw materials used. This grouping
may reflect the differences in raw material usage or the firing
conditions.
6. A detailed petrographic study of other pottery prod-
ucts (dolia, tegula, spica) and mortar (from water tanks) was
made to enable a complex evaluation of raw material usage.
7. The petrographic study was completed with a micro-
mineralogical investigation of accessory minerals for rep-
resentative amphorae and field samples of possible raw
materials.
8. Later on, an electron-microscope (SEM-EDX) and
chemical analysis were used to support, complete and make
the investigation more complex.
In a recent study, we showed the results obtained only
by the micropetrography of Fažana amphorae and thus
we applied only the first five points of our newly proposed
agenda. The results of comprehensive archaeometric research
will be presented in the near future.
For the current detailed petrographic investigations, thin
sections of more than 600 amphorae have been made. More
than 70 different geological outcrops have been visited to
take field samples and to make more than 90 thin sections
from terra rossa and terra rossa soil samples, flysch bedrock
samples, flysch debris flow samples and samples from recent
or sub-recent river and seashore sediments (Fig. 12.1). In the
case of each soft rock sample, two differently prepared small
18.03.19 14:40
 CHAPTER 12 – Micropetrography of the Fažana amphorae (Szakmány – Józsa)
Fig. 12.1 Geological map of Istria with field sampling locations. Terra
rossa appears on the top of limestone (with grey colour). Red
spots: terra rossa, yellow spots: flysch, blue spots: recent sedi-
ments. The basic map is from Živkovic and Bogner (2006),
slightly modified.
bricks (2 × 3 × 6 cm) were formed for thin section studies.
One was prepared only by drying, the other – after drying –
by firing at approximately 750 o C. For petrographic descrip-
tions, a Nikon OPTIPHOT2-POL polarizing microscope
was used. Photos were made using a Nikon DS Fi1 camera
and adjusted using the NIS Elements program.
RESULTS
BRIEF THIN SECTION MICROPETROGRAPHY
OF AMPHORAE
The plastic material of most typical Fažana amphorae is red-
dish brown, rarely yellow, slightly or non-oriented, usually
anisotropic. Originally more or less compact but usually has
variable microporosity.
The proportion of non-plastic components is variable,
the non-carbonatics are generally low (max 10 %, occasion-
ally more), while the amount of calcareous components has a
widespread range from 0 % to approximately 30 %. The non-
plastic material is well sorted, most of the grains are angu-
lar and less than 100 µm in size. Mono- and polycrystalline
BEZECZKY_Text.indd 189
189
quartz dominate, K-feldspars (orthoclase, microcline and
sanidine), plagioclase, different chert and/or silicified matrix
of acid volcanites and small crystalline rock fragments are
rare, micas have variable quantities (muscovite is more com-
mon than biotite). The most common accessory minerals
are garnet, hornblende, tourmaline, epidote, zoisite, rutile,
Cr-spinel and zircon. They are usually few and small sized.
Micrite and sparite appear in variable quantities, calcareous
biogenic remnants (more forams and fewer molluscs) are
frequent. Siliceous biogenic remnants (more opalic, fewer
chalcedonic sponge spicules) are typical but few. Different
sedimentary rock fragments (sandstone, siltstone, claystone,
limestone) are larger sized (0.3-2 mm) and more rounded,
but rare.
DETAILED THIN SECTION MICROPETROGRA-
PHY OF POSSIBLE RAW MATERIALS
In the wider surroundings of the Fažana amphora workshop
in Istria three types of possible raw materials for amphorae
appear in considerable amounts on the surface.
1 TERRA ROSSA
The terra rossa appears as sedimentary cover on Pre-Pleis-
tocene karst surfaces of mostly Mesozoic limestones in the
Mediterranean region. In the study area terra rossa and terra
rossa soil can be found in different thicknesses (from a few
centimetres to more than ten metres, with an average of 1
m) on the surface in vast areas in southern Istria (Red Istria)
and sporadically in northwestern Istria, in the surroundings
of Koper bay5 (Fig. 12.1). Terra rossa and terra rossa soil sam-
ples were collected and analysed from 28 places in southern
Istria.
Macroscopically terra rossa and terra rossa soil are very
dense, massive, pore-free, homogeneous, red clayey sedi-
ments (Fig. 12.2). Often contains a few large fragments
(around 0.1-2 cm) of underlying limestone, which are easily
visible to the naked eye.
With a polarising microscope, two types are distin-
guished. The pure terra rossa is a bright red-coloured homo-
geneous clay (with submicroscopic grain size), often with a
slight, irregularly striped structure. There is no clastic mate-
rial in it. This type of terra rossa is rare in southern Istria.
The mixed terra rossa (or terra rossa soil) has a homogeneous
bright red, very fine-grained (submicroscopic) matrix with
5
Durn et al. 1999; Durn 2003.
18.03.19 14:40
 190 MICROPETROGRAPHY
a considerable amount of uniformly distributed siliciclas-
tic material (Pl. 35.1). The well-sorted, angular monomict
siliciclastics have a fine grain size (40-80 μm). Non-plastics
of mixed terra rossa mostly consist of quartz (dominantly
with wavy extinction, few are volcanic); feldspars and micas
are rare. Accessory minerals appear remarkably frequently.
There are no calcareous materials (except for the very large,
micritic limestone fragments, with typical sparitic veinlets
and fossil infillings, derived from the underlying Mesozoic
limestone bedrock layers mentioned above) and there are no
fossils in it.
2 FLYSCH
Flysch is a deep marine sedimentary rock series. In northern
Istria (Grey Istria) flysch is of the Cenozoic age and consists
of many different sedimentary rock types with variable ratios
of three main types of materials: the siliciclastics, the carbon-
ates and the clay and claystone6 (Fig. 12.3). More than 35
samples were collected from different rock types of flysch
series, including bedrock outcrops and underlying debris
cones. Components of fluvial sediments from northern
Istrian rivers which originated from flysch were also studied.
Basically flysch has three different types here:
In the siliciclastic rock types, grains are angular, medium
sorted with 40-200 μm grain size. Most of the grains are
quartz with wavy extinction. Feldspars are not rare; mica
usually is. Accessory minerals are markedly few, but the
appearance of Cr-spinel in the northernmost Istrian sam-
ples is characteristic. Among siliciclastics, two characteristic
types of microcrystalline quartz rock fragments appear in
small quantities. The chert of sedimentary origin is very fine-
grained, homogeneous and colourless, with a grain size of
less than 10 μm (Pl. 35.2). The volcanogenic type is similarly
fine-grained, but less homogeneous and has reddish colour-
ing. A few fossil chalcedony sponge spicules are also charac-
teristic (Pl. 35.3).
Calcareous flysch rock types are also frequent. One type
is the pure, homogeneous, clay-bearing, micritic limestone.
The other main type is the calcareous sandstone, usually
consisting of a few siliciclastics and many different kinds of
angular micritic and sparitic grains and different calcareous
fossil fragments (Pl. 35.4).
The third main rock type of the flysch series is the mainly
homogeneous greyish clay and claystone with some micrite
content. Both calcareous and clayey rock types often contain
6
Gulam et al. 2014; Mikes et al. 2006.
BEZECZKY_Text.indd 190
Fig. 12.2 Terra rossa (reddish brown sediment) covers the Mesozoic
limestone (white rock) in the coastal zone of Istria. Recent
shallow marine sediment (grey sand) is visible on the lower left
side of the picture.
variable quantities of different types of globigerinida fora-
minifera (thick-shelled or thin-shelled, occasionally filled
with opaque minerals), which are very characteristic for
north Istrian Cenozoic flysch (Pl. 35.5).
3 RECENT SEDIMENTS
Thin sections, made of both dried and fired recent soft sedi-
ments from different marine and river environments in Istria
have also been studied. Samples collected from the bank
of the lower course and influx of different rivers contain a
strongly reworked mixture of different varieties of two main
rock types (namely, in northern Istria mainly the mixture of
rocks of flysch series and in southern Istria mainly the terra
rossa with a small amount of Mesozoic limestone fragments),
eroded from the gathering ground of the given river.
Recent marine sediments partly consist of recent skel-
eton fragments of living beings of marine origin mixed with
variable quantities of the previously mentioned terrigenous
material in the same territorial distribution (in northern
Istria, mainly the mixture of rocks of the flysch series and in
southern Istria mainly the terra rossa with a small amount of
Mesozoic limestone fragments). These sediments have been
collected from three different Istrian coastal environments.
In the description below, we concentrate only on the com-
ponents with marine origins, as the terrigenous rocks were
described in the previous paragraph.
1) The coarse-grained calcareous sand was sampled from
the tidal level of the bay, subject to strong waves, of Salina
on Brijuni Island, southern Istria. This material is medium
18.03.19 14:40
 CHAPTER 12 – Micropetrography of the Fažana amphorae (Szakmány – Józsa)
Fig. 12.3 Outcrop of a flysch rock series
from the interior of Istria, north
of Zajci (IST 5).
sorted with an average grain size of 0.5-1 mm. It consists
mainly of the skeleton fragments of recent marine animals,
among others large empty shells of mollusc and foramin-
ifera (Pl. 35.6), as well as Mesozoic micritic limestone grains
(larger pieces often with sparitic veinlets). No fine-grained
matrix is present.
2) The sandy silt was collected from the bottom of a shal-
low sea on rocky beaches. In addition to the aforementioned
coarse biogenic fragments, these sediments contain variable
amounts of finer-grained calcareous and siliciclastic grains
from inland territories.
3) The muddy silt was taken from the tidal or supratidal
zone of different, very flat and well-protected bays near
Koper and on Brijuni Island. Flat bays with lots of very fine-
grained sediments near Koper were mostly destroyed by the
new building activities, causing sampling difficulties. But not
so in Brijuni Island, where the bays are only slightly disturbed
by human activities. These samples consist of few but usu-
ally larger-sized marine biogenic remnants in a fine-grained,
silty, mud-rich mixture of main source rock types (flysch in
northern Istria and terra rossa with Mesozoic limestone frag-
ments on Brijuni Island, southern Istria). The homogeneity
of the material collected depends on the distance of the place
of extraction from the source territory of the accumulated
rock types.
In the recent marine sediments some characteristic fea-
tures were detected. In the sand-rich samples, some of the
larger calcareous biogenic fragments and limestone grains
BEZECZKY_Text.indd 191
191
show traces of bioerosion (boring traces of piddock and
sponge) (Pl. 35.7). In silt-bearing samples, opalic sponge
spicule fragments appear in two main forms. One has a
spiny shape (Pl. 35.8), the other has a spherical (sterraster)
appearance (Pl. 36.1). They appear consistently but only in
small quantities, and at the same time, the spherical type
always appears in a much smaller proportion. In addition,
the fine-grained, silty mud samples from the flat coasts of
different small bays of the Gulf of Trieste may contain both
components mentioned above (fragments of recent skel-
etons, Mesozoic limestone and terra rossa). Beyond that,
some of these samples contain very long, arched sections of
thin-shelled clams as well (Pl. 36.2). All these components
are embedded in a calcareous silty mud matrix originating
from flysch, which has a reddish-brown colour after firing at
750 oC (Pl. 36.3).
DETAILED COMPARATIVE THIN SECTION
MICROPETROGRAPHY OF FAŽANA
AMPHORAE
In the course of our detailed comparative petrographic
research, we found a significant number of components
and textural features in Fažana amphorae, which show
unambiguous similarities with those materials detected and
described in the potential raw materials for Fažana ampho-
rae (terra rossa and terra rossa soils, flysch rock types, recent
sediments) collected from the field in Istria (see Fig. 12.1).
18.03.19 14:40
 192 MICROPETROGRAPHY
Most of them are easily recognizable and simply identifiable
by petrographic microscope.
A brief thin section micropetrography of Fažana ampho-
rae was presented at the beginning of this chapter. In this
section, we present only those components found in thin
sections of Castrum amphorae, which have clear relevancy
in raw material identification. Characteristic features are col-
lected in subgroups and each of them is shown separately.
1 GROUNDMASS (PLASTIC MATERIAL)
- The reddish-brown colour of the groundmass of Castrum
amphorae (Pls. 36.8, 37.1, 37.3, 37.5, 37.6 and 37.8) dif-
fers from the bright red colour of the groundmass of fired
terra rossa (Pl. 35/1), and is more similar to the colour of
fired silty mud originating from flysch (Pl. 36.3).
- The groundmass of some amphorae contained fine-
grained micrite in its raw material, which due to higher
firing temperatures may have become light yellow and an
optically isotropic feature. Terra rossa does not contain
any fine-grained calcareous material, so this colour indi-
cates the presence of a considerable amount of micrite-
rich components in the raw material of Fažana amphorae.
In some varieties of fine-grained flysch rock types studied
a considerable amount of micrite was detected.
2 NON-PLASTIC MATERIALS
- The amount of non-plastic materials in Fažana amphorae
is highly variable, but in most of the cases, it is less than
that in the terra rossa soil, moreover, in many samples it
is almost absent, as in the fine-grained flysch-derived rock
types.
- There are two main representative types, the siliciclastic
and the calcareous clasts. The ratio of siliciclastic and cal-
careous fragments is extremely variable, but the calcareous
material usually exceeds the amount of siliciclastics and in
some cases becomes exclusive, in other words, the silici-
clastics appear in small or very small quantities (Pl. 36.4).
These features are different from those of mixed terra
rossa, which contains larger amounts of siliciclastic mate-
rial (Pl. 36.5) and there is no calcareous material in it.
However, all these features can be found in different rock
types from Istrian flysch sequences.
2.1 MONOMINERALIC COMPONENTS
- Microquartz fragments are angular and have a similar
grain size to other siliciclastics (Pl. 36.6). The quantity
BEZECZKY_Text.indd 192
in amphorae is usually low, but in some samples may
increase up to c. 30  % of the total siliciclastics. Two types
have been distinguished.
1) The first one is colourless, clear and homogenous with a
uniform, very fine grain size. Most likely, these grains are
fragments of cherts.
2) The other one also has a very fine, but slightly variable
grain size. Its colour is light red because of disseminated
hematite and limonite. Most likely, these grains are frag-
ments of the groundmass of acidic volcanite rocks.
These components are frequent in siliciclastic-rich flysch
rock types (Pl. 35.2), and very rare or absent from the
terra rossa. In some amphora samples both microquartz
grain types have been detected as components of flysch-
type rock inclusions (see below).
- The accessory minerals can be good indicators for raw
material discrimination.
1) The total quantity of accessory minerals in amphorae is
usually low. We were able to observe the same feature in
the case of flysch rock types, while all terra rossa samples
contain a lot of accessories. The difference is usually
observable even with a polarising microscope at higher
magnification.
2) The Cr-spinel, as the best marker of accessories can be
detected by polarising microscope only in a few amphora
samples because of its rarity and comparatively small
grain size. The possible appearance could still be interest-
ing, since this accessory mineral is characteristic especially
for those flysch rock types which appear in the northwest-
ern part of Istria and the surroundings. In some Fažana
amphora samples, Cr-spinel grains have been detected as
components of flysch-type rock inclusions (see below).
Terra rossa does not contain any Cr-spinel.
- Micrite and sparite fragments are very characteristic for
most of the Fažana amphorae, in some cases forming the
dominant part of its non-plastic material (Pls. 36/4 and
36/7). Their grain size varies across a slightly wider inter-
val range than those of siliciclastics. In the samples of high
fired amphorae, carbonate grains can be recognised only
in the form of pores. These kinds of carbonates are abso-
lutely missing from terra rossa and typical for calcareous
flysch rock types (Pls. 35/2, 35/3 and 35/4).
2.2 INDIVIDUAL FOSSILS
- In the Fažana amphorae, the individual skeletons of dif-
ferent kinds of globigerinida are typical and appear in
large quantities. The most widespread types are the fossil
globigerinida. In the amphorae fired at lower tempera-
18.03.19 14:40
 CHAPTER 12 – Micropetrography of the Fažana amphorae (Szakmány – Józsa)
tures, most globigerinida are easily recognizable on the
basis of microsparitic infillings (Pl. 36.7). Among these
skeletons, both thin- and thick-shelled can be distin-
guished. In some cases, a very characteristic opaque fill-
ing is visible (Pl. 37.1). In many amphorae the original
calcareous material of globigerinida is often missing due
to the effects of higher firing temperatures, and the glo-
bigerinida origin of the pores could be confirmed only on
the basis of their shape and size. Different forms of fossils
similar to all these globigerinida remnants are common
constituents of the Cenozoic flysch of northern Istria
(Pl. 35.5). These fossils have a decisive importance since
they appear in similar forms as components of flysch-
type rock inclusions (discussed below in detail) in some
amphora samples. Moreover, the terra rossa samples do
not contain any remnants of living beings.
- Other types of calcareous foraminifera and mollusc skel-
eton fragments are rare, but characteristic for Fažana
amphorae (Pl. 36.8). Based on our research, these frag-
ments look very similar to certain fossil components
appearing in field flysch samples collected in northern
Istria.
- Siliceous fossils are rare in Fažana amphorae, and are rep-
resented only by sponge spicules consisting of chalcedony
with a slightly radial structure (Pl. 37.2). Despite its rarity,
it is a good indicator, as same fossils appear – also in small
quantities but characteristically – in the siliciclastic-rich
rock types of Istrian flysch (Pl. 35.3). In some amphora
samples chalcedony sponge spicules have been detected
as a component of flysch-type rock inclusions (see later).
2.3 RECENT LIVING ORGANISM REMNANTS
- In many Fažana amphora samples, very fine-grained cal-
careous skeletons of different kinds of foraminiferas with
empty chamberlets have been detected (Pl. 37.3). Among
these remnants there are characteristic types, very similar
to the recent foraminiferas observed in large quantities in
recent coarser-grained sediment samples collected from
the seashore of southern Istria (Pl. 35.6).
- In several amphorae, larger sized mollusc skeleton frag-
ments with empty shells have been observed. Similar skel-
etons have been detected in the Istrian seashore sediment
samples. Among them, there is a very characteristic type,
which has a thin, long-arched section. (Pl. 37.4) It can be
found in silty mud coastal sediments in Koper Bay, north-
ern Istria (Pl. 36.2).
- Opalic sponge spicules are very typical and widespread
components of Fažana amphorae, though they usually
BEZECZKY_Text.indd 193
193
appear only in small quantities. Two main types were
observed. One has an elongated spiny shape (Pl. 37.5) and
always appears in a considerably higher proportion than
the other, which has a spherical shape and radial structure
(sterraster) (Pl. 37.6). Both types of spicules commonly
appear in a small quantity and in similar proportions in
all recent shallow marine silty sediments collected from
the sea shore of Istria (Pls. 35.8 and 36.1).
- Recent calcareous skeleton fragments in the amphorae
are easily recognizable if they show traces of bioerosion
(Pl. 37.7). Similar traces were detected in recent biogenic
fragments in coarse-grained sediments collected from the
recent seashore of Istria (Pl. 35.7).
2.4 COMPOSITE SEDIMENTARY ROCK
FRAGMENTS
Although rock fragments in the amphorae occasionally
appear in small quantities and in variable sizes, they consti-
tute important evidence for an indication of the origin of the
raw materials used for amphora production.
- Claystone inclusions in amphorae are homogenous and
usually appear in a larger size, rounded shape and have
bright red or mainly brownish-red, reddish-brown and
brown colours. Two types were detected 1) The bright
red type claystone has an exclusive rare variety, free of any
non-plastic clasts. This kind of rock inclusion shows fea-
tures typical for the “pure” terra rossa, described above.
The other bright red type claystone is less rare. Its ground-
mass is similar to pure terra rossa, but contains well sorted
fine-grained siliciclastic non-plastics and a lot of accessory
minerals (Pls. 37.2 and 37.8). Its appearance absolutely
matches the properties of the “mixed” terra rossa samples
(Pl. 35.1) from Istria. 2). The variable brownish-coloured
claystone inclusions often contain different types of glo-
bigerinida, similar to the ones described above as indi-
vidual fossils in amphorae. These rocks obviously indicate
the clayey rock types of Cenozoic flysch series in the raw
material of Fažana amphorae.
- Sandstone and siltstone whole rock inclusions usually
appear as rounded large fragments and have a very diverse
appearance in Fažana amphorae. Clastic components are
angular and well sorted, the proportion of calcareous-
siliciclastic material is variable. In the coarser-grained
varieties, many kinds of typical components (micro-
quartz, Cr-spinel, chalcedony sponge spicule, micrite and
sparite grains, different kinds of calcareous globigerinida
and other calcareous foraminifera, distinct molluscs)
18.03.19 14:40
 194 MICROPETROGRAPHY
described above as monomineralic components and indi-
vidual fossils have been observed. The accessory mineral
content is low. Most of these rock types with all these
properties and components can be found in the field
samples collected from different flysch series in northern
Istria (Pls. 35.2, 35.3 and 35.4).
- Micrite-rich rock fragments appear in two types. One
is the clay and/or silt-bearing micritic limestone, which
often has a layered character and mostly globigerinida
fossil content similar to those which appear in different
flysch rock types. The other type is the micritic limestone
with sparitic infillings, which appears relatively rarely
in Fažana amphorae. They are usually very large in size,
well rounded and rarely contain strongly re-crystallized
microfossils. Similar Mesozoic limestones can be found
first mainly in southern Istria as the bedrock of terra
rossa, secondly sporadically as fragments in terra rossa,
and thirdly in the recent sea shore sediments collected in
Istria close to the territories with Mesozoic bedrock.
In the finer grained groundmass of certain amphorae,
larger grains of Mesozoic limestones, recent skeleton frag-
ments and occasionally a few terra rossa grains jointly arrange
in dense, long, ribbon-like groups. These amphora samples
have a hiatal texture. This material is similar to components
detected in south Istrian recent coastal sediment.
DISCUSSION AND CONCLUSION
Based on our petrographic results we can declare that the
amphorae found at Castrum in Brijuni are typical Fažana
Dressel 6B type amphorae. Consequently, the petrographic
evaluation of Castrum amphorae can be integrated with or
even based on the petrographic knowledge obtained by the
investigation of all amphorae manufactured in Fažana in the
period under study.
Earlier researchers found – based mainly on micro-
mineralogical investigations – that the main raw material
of Fažana amphorae was the terra rossa, to which a small
amount of recent marine material had been added.7
During our research, three possible main types of raw
materials from Istria were examined. Namely, the two sub-
types (pure and mixed) of terra rossa, the variable rock types
(containing sandy, silty, clayey and calcareous components)
of flysch sequences and the recent sediments from different
7
Mange – Bezeczky 2006; Mange – Bezeczky 2007;
Bezeczky – Mange 2009.
BEZECZKY_Text.indd 194
environments (debris flow, fluvial, shallow marine) contain-
ing terrestrial and in the latter case materials of recent marine
origin.
The pure terra rossa did not play any role in forming the
raw material of Fažana amphorae, since its fragments are
missing from the amphora samples examined. This is con-
sistent with the fact that this rock variety is very rare in sub-
aerial outcrops and in different recent debris in Istria.
In the Fažana amphora samples we could frequently see
just a small amount of terra rossa soil whole rock fragments,
also a small amount of accessory minerals and a huge quan-
tity of calcareous fragments. These observations show that
the terra rossa soil was often present in the raw material of
the Fažana amphorae, but always only in a small quantity.
Based on these observations, we can conclude that the main
raw material of Fažana amphorae was not the mixed terra
rossa. If we take into account also that terra rossa soil is avail-
able in large quantities on the surface in the whole territory
of southern Istria (Red Istria) (thus in Fažana too, where
amphorae were manufactured) and in a smaller surface area
in northern Istria, we can conclude that terra rossa soil itself
was not a useful basic material for amphora production.
Many of the features observed only in different rock types
of Istrian flysch and described in detail above could be found
in the thin sections of most of the Fažana amphora samples.
Moreover, the vast majority of amphora samples consist
almost exclusively of flysch-derived materials, and obviously
contain only few remnants of other sources (mainly marine,
rarely terra rossa and fragments derived from Mesozoic lime-
stone). In addition, in several amphora samples, often even
within one sample, many different whole rock fragments
assuredly derived from flysch rock series could be detected.
Based on these detailed and diverse petrographic observa-
tions we can conclude that the only main source of the raw
materials for Fažana amphora, thus also for the Castrum
amphorae, was the Istrian flysch, which was available in a
huge quantity on the surface only in the northern part of
Istria (Grey Istria), quite far from Fažana where the ampho-
rae were produced.
The presence of recent marine sediments in the raw mate-
rial of the amphorae can be recognized by means of com-
ponents of marine origin (opalic spicules, boring sponge
traces, recent marine skeletons). The quantitative ratio of
these components to the terra rossa or flysch-derived silty
mud eroded and transported into the sea from the inland is
variable and depends mainly on the type (flat or rocky) and
geographical position (northern or southern Istria) of the
coast. The easily recognizable special fragments of marine
18.03.19 14:40
 CHAPTER 12 – Micropetrography of the Fažana amphorae (Szakmány – Józsa)
Fig. 12.4 Raw material transported from Trieste bay (Bezeczky 2015).
origin, most prominently the opalic sponge spicules, appear
in almost all samples of Fažana amphorae, but usually only
in small quantities. Consequently, recent marine sediment
must constantly and consistently have entered the raw mate-
rial of Fažana amphora but the portion of its components of
marine origin was always small in quantity.
As a first summary of the above discussion we can reli-
ably reveal that the fundamental mass of the raw material
of typical amphorae manufactured in Fažana and excavated
in Castrum villa has to derive from north Istrian flysch. In
this main raw material, components of marine origin appear
commonly but only in small quantities, and debris of terra
rossa is usually rare and in a limited quantity.
Beyond that, by further interpretation based on our
detailed observations, more exact conclusions can be drawn.
In the next step, we have to find out more precisely the kind
of material and from where it was extracted.
BEZECZKY_Text.indd 195
195
Flysch material available in the outcrops or in the debris
apron could not be used directly for amphora produc-
tion, because in these kinds of debris there are significant
amounts of large hard rock fragments and these fragments
are rare in the amphorae. It is important to note that the
north Istrian rivers cross large flysch-covered territories. The
channel deposit of these rivers consists of huge quantities
of a soft homogenous detritus mixture, originating mainly
from flysch rock types and free of large hard rock fragments.
Moreover, the fast erosion rate of the flysch leads to the re-
deposition and fast accumulation of fluvial sediments into
the shallow bays at the influx of north Istrian rivers. In these
shallow marine environments, natural immixing of a huge
mass of mostly homogenous soft flysch mash and a few
characteristic marine components (skeleton fragments of
recently living beings) took and still takes place. In addition,
on the drainage area of such a river or even near to the shores
of such a bay, Mesozoic limestone units covered by terra
18.03.19 14:40
 196 MICROPETROGRAPHY
rossa may have existed on the surface. In such circumstances,
a lower volume of terra rossa detritus may have also eroded
and been transported into the bay filled with much flysch
mash and few recent biogenic remnants.
As a final result of these sedimentary processes, in the
north Istrian shallow marine bays connected with river
influxes (Mirna, Piran, Koper, Rio di Ospo) in the wider
coastal area of the Gulf of Trieste we would expect to find
fine-grained, soft and homogenous sediments. These special
sediments could have formed through the natural mixing
of a dominant amount of flysch mash, a small amount of
marine originated skeleton fragments, which appear con-
stantly, minimal barely detected fine debris of terra rossa,
and occasional larger fragments of Mesozoic limestones,
flysch rock types and terra rossa.
We have detected and described exactly the same compo-
nents and ratios in most of the Fažana amphorae. Moreover,
in the fine-grained coastal sediments collected from the
aforementioned north Istrian bays we were able to find simi-
lar components in very similar proportions8. Furthermore,
the long, arched sections of the very thin-shelled clams of
recent marine origin, observed in the matrix derived from
flysch in both the Koper bay marine sediment and in the
thin sections of Castrum amphora, means that the raw mate-
rial was used after excavation without any serious adjustment
(mixing and kneading). The texture of the amphorae and the
fired sediment also appeared to be very similar.
To find the most probable solution for the final ques-
tions connected with the raw material of Fažana amphorae,
the results of geological analyses were assessed and com-
bined with aspects of other activities (economic, mining,
shipping, logistical, pottery, etc.) as well. We think that the
most important aspects to consider are the quality of the
8
Józsa et al. 2016.
BEZECZKY_Text.indd 196
raw material, the simplicity of extraction, the distance and
mode of transport and the method of preparation of the raw
material. All these aspects are connected and/or depend on
geological knowledge.
The quality of the shallow marine fine-grained homog-
enous material of the Gulf of Trieste, according to the
experimental results of Véninger9, could be suitable enough
by itself to produce such good quality large-scale pottery as
the amphorae. Here they could immediately extract such
a material, which did not require any more additives. The
bays here were the closest places to Fažana where a suffi-
ciently good quality of raw material could be found. The
place of extraction was most probably located on the coast,
so that it could be loaded directly into the ships and easily
transported away.
SUMMARY
On the basis of our research we can surmise that the raw
material of Fažana amphorae was most probably extracted
from the coastal zone in one of the small bays of the Gulf of
Trieste in the reach between Umag and Trieste. For most of
the amphorae studied this was the only raw material used.
After the extraction, the flysch-based raw material was
transported 60-70 km by sea from the Gulf of Trieste to
Fažana. The exact composition, and consequently the qual-
ity of manufactured amphorae, depended on which bay and
exactly what part of the pit the raw material was extracted
from. If the responsible potter was not completely satisfied
with the clay, he may have optimized it by adding material
available in Fažana. Occasionally, or at certain times, the
potter could add some coarser-grained marine sediment for
leaning, or terra rossa mud for fattening.
9
Véninger 2016.
18.03.19 14:40
 PLATE 35

1 2

3 4

5 6

7 8

1. Polarizing microscopic photo of fired sample of mixed terra rossa from SW of Kavran, Istria (IST 12, PPL, picture width 0.6 mm). – 2. Polarizing
microscopic photo of angular chert grain (granular gray grain in the middle) in calcareous sandstone. Field sample from flysch series near Seča, at Piran
bay, Slovenia (IST 64b, XPL, picture width 0.3 mm). – 3. Polarizing microscopic photo of cross section of chalcedony sponge spicula fragment (grayish
sphere on the right) in calcareous sandstone. Field sample from flysch series near Seča, at Piran bay, Slovenia (IST 64b, XPL, picture width 0.3 mm). –
4. Polarizing microscopic photo of calcareous sandstone with different fossils. Field sample from flysch series near Seča, at Piran bay, Slovenia (IST 64b,
PPL, picture width 0.6 mm). – 5. Polarizing microscopic photo of micritic claystone with different characteristic fossils. Field sample from flysch series
near Katun Trvižki, Istria (IST 31a, PPL, picture width 0.6 mm). – 6. Polarizing microscopic photo of characteristic finegrained brown calcareous empty
skeleton of a recent foraminifera. Recent shallow marine sandy silt sample from Fažana, Istria (IST 16d, PPL, picture width 0.6 mm). – 7. Polarizing
microscopic photo of calcareous skeleton fragment of recent mollusc with traces of boring sponge. Recent shallow marine sandy silt sample from Fažana,
Istria (IST 17, PPL, picture width 0.6 mm). – 8. Polarizing microscopic photo of recent opalic sponge spicula of spiny form (in the middle). Recent
shallow marine sandy silt sample from Fažana, Istria (IST 17, PPL, picture width 0.6 mm).

BEZECZKY_Text.indd 35 18.03.19 14:40

 PLATE 36

1 2

3 4

5 6

7 8

1. Polarizing microscopic photo of recent opalic sponge spicula of spheric form (sterraster). Recent shallow marine sandy silt sample from Fažana, Istria
(IST 16, PPL, picture width 0.3 mm). – 2. Polarizing microscopic photo of thin long ached section of a shell in recent silty mud sediment from the
flat coast of Koper bay (IST 67, PPL, picture width 1.2 mm). – 3. Polarizing microscopic photo of a fired flysch originated silty mad from flat coast of
Koper bay with thin arched shell fragment (IST 68, PPL, picture width 0.6 mm). – 4. Polarizing microscopic photo of a typical amphora from
Castrum. Siliciclastic content is low, larger calcareous grains are present (Castrum 13, XPL, picture width 1.2 mm). – 5. Polarizing microscopic photo of
a typical terra rossa from a quarry close to Kavran. The sample fired in 750 oC has high siliciclast content (IST 12, XPL, picture width 1.2 mm). – 6. Po-
larizing microscopic photo of angular chert grain (granular gray grain on the bottom left) in a Castrum amphora (Castrum 14, XPL, picture
width 0.3 mm). – 7. Polarizing microscopic photo of microsparite filled globigerinida foraminifera in Castrum amphora (Castrum 18, PPL, picture
width 0.6 mm). – 8. Polarizing microscopic photo of micritic grains and mollusc skeleton fragment in a Castrum amphora (Castrum 12, XPL, picture
width 0.3 mm).

BEZECZKY_Text.indd 36 18.03.19 14:40

 PLATE 37

1 2

3 4

5 6

7 8

1. Polarizing microscopic photo of opaque filled globigerinida foraminifera (black in the middle) in Castrum amphora (Castrum 16, PPL, picture
width 0.3 mm). – 2. Polarizing microscopic photo of fossilic sponge spicula consisting of chalcedony (grayish spheric grain in the left down quarter) in
a Castrum amphora. On the right side rounded terra rossa inclusion (Castrum 13, XPL, picture width 0.3 mm). – 3. Polarizing microscopic photo of
empty shelled recent foraminifera (left down corner) and cross section of opalic sponge spicula (right upper corner) in a Castrum amphora (Castrum
18, PPL, picture width 0.6 mm). – 4. Polarizing microscopic photo of section of thin long arched shell in a Castrum amphora (Castrum 14, PPL,
picture width 1.2 mm). – 5. Polarizing microscopic photo of elongated opalic sponge spicula in a Castrum amphora (Castrum 16, PPL, picture width
0.3 mm). – 6. Polarizing microscopic photo of spheric opalic sponge spicula (sterraster) in a Castrum amphora (Castrum 16, PPL, picture width
0.3 mm). – 7. Polarizing microscopic photo of recent marine calcareous living being fragment in Castrum amphora with characteristic traces of bio-
erosion caused by boring sponges (Castrum 12, PPL, picture width 0.3 mm). – 8. Polarizing microscopic photo of well rounded terra rossa soil rock
fragment (in the middle) in a Castrum amphora (Castrum 22, PPL, picture width 0.6 mm).

BEZECZKY_Text.indd 37 18.03.19 14:40

BEZECZKY_Text.indd 38 18.03.19 14:40

13 SUMM ARY
Tamás Bezeczky
As every individual chapter on the amphorae contains a
detailed summary, we will only provide an overview here.
Taking all of the amphorae into consideration, our ear-
lier hypotheses have been corroborated. It is clear that, in
the Early Roman period, the Castrum villa was part of an
important agricultural unit which produced and exported
olive oil. They imported little wine and fish sauces. The situ-
ation changed in the Mid- and Late Roman periods. They
produced little oil, wine and fish sauces, and the inhabitants
thus consumed imported foodstuffs. This was particularly
characteristic of the Late Roman period, when the villa’s
function changed. The Castrum villa played an important
part in northern Adriatic trade.
INTRODUCTION
Aquileia was founded at the beginning of the second cen-
tury B.C. and became an important military and commer-
cial center. From then on, the Roman military controlled
the population of the northern Adriatic coastline, putting
an end to the wars between the various Illyrian and Istrian
tribes. During Caesar’s campaign, Tergeste and Pola were
granted the rank of colonia, which ushered in a period of
rapid economic development in the Istrian peninsula. Octa-
vianus/Augustus became acquainted with the area between
the Drava and Sava rivers during the war against the Iapodes
(35-31 B.C.). Augustus and his inner circle were generally
known to be interested in Istria. They acquired land there
and soon annexed the entire peninsula. It was the Romans
who introduced the olive plantations and the vineyards on
the western shore. From Pola to Tergeste traces of several vil-
las with olive-producing facilities have been discovered, but
their owners could not be identified. However, there are two
known amphora workshops in the region. One figlina has
been found near Poreč (Parentium) in the bay of Loron. The
other is in the south in Fažana, opposite the Brijuni Islands.
During the first period of the workshop, the owners were
successive members of an influential Roman family, the Lae-
canii. Three villas on the island of Brijuni have been identi-
BEZECZKY_Text.indd 201
201
fied (Verige – Val Catena, Kolci – Monte Collisi, Castrum).
The amphora stamps suggest that the villas were in the pos-
session of the Laecanius family.
The huge quantity of amphorae allows us to investigate
the trade of the Castrum villa.
BEFORE THE MIDDLE OF THE FIRST
CENTURY B.C.
There is a mysterious late Republican Roman villa on the
grounds of the Castrum villa, with a few finds, including
some Lamboglia 2 amphorae. These Adriatic amphorae
from the first century B.C. represent the earliest indication
of wine imported from Italy. The earliest phase of the Cas-
trum villa ended before the middle of the first century B.C.
THE MIDDLE OF THE FIRST CENTURY B.C.
TO THE BEGINNING OF THE SECOND
CENTURY A.D.
Many more Lamboglia 2 amphorae belong to the second,
U-shaped villa built after the middle of the first century
B.C. Some of the tegulae used in the construction of the sec-
ond villa were imported from northern Italy, supplementing
local production.1 This second villa was constructed and
used by the members of the Laecanius family. The design
of the building, the cellar and the tools for oil production
(presses, lacus, dolia) was identical to that of the other vil-
las in Brijuni (Verige – Val Catena, Kolci – Monte Collisi).
The amphorae and various ceramics (stoppers, tiles, spicae,
heating pipes and clay lamps) were made on the mainland
across from Brijuni, in the Fažana workshop. There are some
unstamped Dressel ante 6B amphorae from the early phase
of the villa. Even after petrological analysis, it is impossible to
determine whether they were imported or if they were con-
1
See Berni Millet in this volume: PANSAE·VIBI, A·FAESONI·AF,
C·CEIONI·MAXI, C·FLAVI and the local C.LAECANI.P.F. and
LAEC.
18.03.19 14:40
 202 SUMMARY
nected to local olive oil production. However, one Dressel
ante 6B amphora bearing the C·LAECANI·P·F stamp was
found near Venice. This makes it clear that the Laecanius
family continued to produce these early Dressel 6B ampho-
rae after Caesar’s rule, and that they participated in the pro-
duction and trade of olive oil. There are many indications
that Istrian oil was transported to northern Italy, especially
the settlements of the Po Valley. The most important trade
relations were with Padua and Aquileia, but the amphorae of
the Laecanius family, and, later, the emperors, can be found
in practically every town all the way to Turin. Through
Aquileia, Istrian villas – including in Brijuni – supplied the
new provinces (Raetia, Noricum, and Pannonia) with olive
oil. From the second part of the first century B.C. onwards,
oil was also produced in northern Italy and transported in
Dressel 6B amphorae. Ultimately, oil production in Italy
ceased after the middle of the first century A.D. The start
of oil production at the Dragonera villa can perhaps also be
explained by the fact that the Istrian villas needed to increase
their production. This was the period in which the eastern
border of Pannonia was extended to the Danube, and an
increasing number of soldiers was sent to that province. The
last member of the Laecanius family died without an heir
in 78 A.D. The workshop in Fažana became the property of
Emperor Vespasian. The change in ownership is reflected in
the amphora stamps. The Imperial administration took over
the workshop, the villas, the workers and the people who
were in charge of production. The stamps demonstrate that
oil production was continuous until the reign of Hadrian.
The imported finds from the second Castrum villa include
a number of wine amphorae (Dressel 6A, Dressel 2-4,
Gauloise 2, Rhodian, Knidian, Koan, Dressel 5, Cretan and
Agora F65/66). There was no evidence of the local produc-
tion of fish sauces during this period, but there was a limited
quantity of imports from Hispania (Dressel 7-11, Beltran II
A and II B and Dressel 14 A amphorae). There were some
demonstrable differences between the Castrum villa and the
areas north of Aquileia in terms of trade. While there were
hardly any fish sauces from Hispania in the Castrum villa,
there were large quantities of these fish sauces in the north, at
the trading center in Magdalensberg and in Pannonia.
The villa was probably reconstructed or renovated several
times. This is what the tegulae from the reigns of Tiberius,
Caligula and Nero and other tegulae from the first century
A.D. suggest. There is no information regarding a possible
connection between the Istrian workshops, or the con-
nections they may have had with northern Italian work-
shops. Tegulae from the Loron workshop have been found
BEZECZKY_Text.indd 202
in the Castrum villa, with the stamps SISENNAE and
IMP·AVG·GER. Approximately 60-70 years passed between
the production of the two stamps. How and why these tegu-
lae ended up in the villa is unclear, and it is difficult to say
which of the possible explanations is correct.
THE MIDDLE OF THE SECOND CENTURY TO
THE END OF THE THIRD CENTURY
A new situation arose when Istrian oil production ceased
after the Hadrianic period. The characteristic Dressel 6B
amphorae disappeared from the traditional markets when
Istrian oil production decreased to a minimum. Merchants
filled the gap with imported oil from Hispania Baetica,
transported in Dressel 20 amphorae. They supplied all of the
markets which were formerly supplied by Istrian oil. The lack
of oil was so acute that oil from Baetica was imported even
to the Castrum villa. There is a Dressel 20 stamp which can
be dated to this period (no. 40). Some think that the prob-
lems of Istrian oil production and export are explained by
the Marcomannic wars, which broke out during the reign of
Marcus Aurelius, and which made it impossible to transport
Istrian oil to the provinces. However, the northern provinces
had been using oil from Baetica for a long time by then. Thus,
the reason must have been a change in Istrian oil production.
Baetican amphorae can be found until the end of the third
century, albeit not in large quantities. A Dressel 20 amphora
which can be dated to the second third of the second cen-
tury A.D. and bears a graffito (Asiatici / sec( ) x k(alendas)
mar(tias) Alb( )) is of particular interest.
African amphorae appear in the villa from the middle of
the second century to the middle of the third century (Afri-
cana 1 A/B and Africana II A). Some of these may be con-
nected to the lack of olive oil. African vessels were brought
to the northern Adriatic region far earlier, primarily to Aqui-
leia, as well as a small quantity to Noricum and Pannonia.2
On the one hand, the frozen olive trees had to be replanted.
On the other, the production and export of fish sauces prob-
ably began at this time. New amphora types (Fažana 1 and
2) appeared in the second half of the second century, which
clearly reflects changes in agricultural production. This must
have gone hand in hand with changes in production at the
2
Tripolitana II amphorae were transported from Aquileia to Mag-
dalensberg and Emona. Africana I and Schöne-Mau XL amphorae
were found near the Danube in Ovilava (Noricum). Leptiminus 1
amphorae were found at the Iseum in Savaria, along the Amber Road.
One of the best-known finds in the northern Adriatic region are the
Africana I amphorae found at the Grado shipwreck.
18.03.19 14:40
 CHAPTER 13 – Summary (Bezeczky)
Fažana workshop. We only know the name of one pro-
ducer in this period, although the production of amphorae
restarted in other places, too, such as Loron. The stamp of
M. Aurelius Iustus was found in Fažana, and his altar was
found in the cemetery next to the Castrum villa. His stamp
was on a Fažana 1 amphora. We do not know when and for
how long this amphora type was produced; it must have
been connected to oil production starting up again. Vessels
of this form have not been found outside of Istria.
Wine was initially imported from Italy. Later, wine was
also produced on the island, although it is unclear when
production started. In the late Roman period, only Cassio-
dorus mentions Istrian wine. The number of presses suggests
that a significant quantity of wine was produced. If these
were exported, what vessels were used for this purpose? As
we do not have chronological data, it is impossible to date
the Forlimpopoli wine amphorae precisely. Their origin also
requires further study.
After the end of the second century, new types of wine
amphorae appeared from Asia Minor, from the region of
Pergamum and Ephesus (Agora M 125, M 126, M 240)
and other parts of the Aegean (Agora G 199, Kapitän I and
Kapitän II). A few of the amphora types from the eastern
Mediterranean were also found at the underwater excava-
tions at Val Catena.3 Oriental amphorae are fairly common
in the northern Adriatic region. They are found less fre-
quently in the northern provinces, with the exception of the
Kapitän II amphorae, which are found in greater quantities
along the Pannonian limes.
Starting at the mid/second half of the third century, Afri-
can salted fish and/or wine was imported from Nabeul and
Thaenae (Africana II C and II D amphorae). They are also
found in Aquileia. Wine (?) was also imported in Beltran 68
amphorae from the Gaditan workshop in Baetica starting in
the late third century to early fifth century.
THE FOURTH CENTURY TO THE BEGINNING
OF THE FIFTH CENTURY
Who used the villa? There is no information regarding this
issue after the time of M. Aurelius Iustus. It seems likely that
Istria and Brijuni came under the direct authority of the
vicarius of Milan following the reforms of Emperor Diocle-
tian. The fullonica workshop, which represented an entirely
3
Bloier 2012, Agora M 126, no. 1421, M 240 nos. 1506, 1570 and
1575.
BEZECZKY_Text.indd 203
203
new function of the villa, is presumed to have been estab-
lished during this period?
Olive oil was probably also produced. If oil was imported
from the eastern Mediterranean, it is very likely to have been
transported in either LR 1, LR 2 amphorae and Ephesus 56
amphorae.
We do not have any information regarding local wine
production in this period, either. It is not surprising that
African wine was brought to the villa (Africana III A, III B
and a variant of Keay 1B) from Nabeul. The Africana III is
one of the best-known amphorae in the northern Adriatic
region. The import of Lusitanian fish sauces also started in
this period in Almagro 50 (Keay XVI, Keay XXII), Almagro
51C and Keay XIX (Almagro 51A-B) amphorae.
THE FIRST HALF OF THE FIFTH CENTURY TO
THE FIRST HALF OF THE SIXTH CENTURY
The fifth century saw the construction of the walls protecting
the villa, thus determining the final shape of the settlement.
St. Mary’s Church was built about one hundred meters from
the settlement. The construction of the church presumably
altered the function of the settlement. The products of the
villa remain unknown. However, Cassiodorus tells us that
Istria was considered to produce good quality grains, wine,
oil, oysters and fish at the beginning of the sixth century, and
its wine and oil were exported to Ravenna.4 One wonders
why, if Istria was so rich, it was necessary to import similar
foodstuffs from far away.
There are large quantities of the amphorae of Cilicia, the
Greek mainland, Thasos, Levant (LR 1, LR 2, LR 4) and
amphorae from the Ephesus region (Agora M 307, LR 3 and
Ephesus 56). Late amphorae from other areas (Crete and the
Black Sea) can also be found in Brijuni.
From the first half of the fifth to the first half of the sixth
centuries, a significant quantity of African amphorae were
transported to the settlement. A number of products from
Nabeul: spatheia type 1, Keay 3/5, 35A and B, 55, and 57
were found. South Byzacena types Keay 59/8B, northwest-
ern types Keay 27/36, and unprovenanced type Keay 62Q
were also found.
4
Cassiod. var. 22: “For what Campania is to Rome, Istria is to
Ravenna – a fruitful Province abounding in corn, wine, and oil; so
to speak, the cupboard of the capital.” 24: “We have previously given
orders that Istria should send wine and oil, of which there are abun-
dant crops this year, to the Royal residence at Ravenna.”
18.03.19 14:40
 204
“Paul Reynolds already noted this trend in the whole
Adriatic from the mid-fifth century onwards, explaining
it by the ‘major shipping route to Ravenna’, and the ability
for cities and settlements located along this route to benefit
from these exports”.5
THE MIDDLE OF THE SIXTH CENTURY TO
THE END OF THE SEVENTH CENTURY
In this period, African imports came from the coastal central
Byzacena region in Keay 62 and 61 amphorae from Mok-
nine, Keay 61C from the Salakta hinterland, Keay 8A from
unknown Tunisian Sahel sources, and a miniature spatheia
of type 3C from Nabeul. There were fewer amphorae from
5
Reynolds 2010, 99; see also Auriemma – Quiri 2007, 34.
BEZECZKY_Text.indd 204
SUMMARY
the eastern Mediterranean than in the period before, per-
haps only the Ephesus LR 3, Ephesus 56, Samos Cistern type
and an LRA 13 amphora made on the island of Kos. The
latter bears an Imperial seal.
The Castrum settlement must have had a particularly
important role in the fifth to the seventh centuries. St Peter’s
Church was built in the sixth century, about three hundred
meters from the settlement. It is currently unclear if the Bri-
juni Castrum became a Byzantine military fortification or a
settlement for the civilian population. The fact that the two
churches are outside of the walls suggests that the latter is the
case. “The evidence presented here indicates that there was
undoubtedly a bishop’s see in Brioni which existed for more
than two and a half centuries.”
18.03.19 14:40

13 SAŽETAK
Tamás Bezeczky
Kako se na koncu svakog poglavlja o amforama nalazi iscr-
pan sažetak, na ovom mjestu ponuditi ćemo opći pregled.
Uzimajući u obzir sve primjerke amfora, naše ranije pretpo-
stavke čini se potvrđenim. Jasno je kako je u vrijeme ranog
Rima vila Castrum bila dio važnog poljoprivrednog dobra
koje je proizvodilo i izvozilo maslinovo ulje. Uvoz se sastojao
od manjih količina vina i ribljih umaka. U vrijeme srednjeg i
kasnog Rimskog carstva situacija se promijenila. Proizvodnja
se ograničila na manje količine ulja, vina i ribljeg umaka pa
su stanovnici konzumirali uglavnom uvezenu hranu. Ovo se
posebice odnosi na kasni period kada se funkcija vile u pot-
punosti promijenila. Vila Castrum igrala je bitnu ulogu u
trgovini na području sjevernog Jadrana.
UVOD
Akvileja je osnovana početkom drugog stoljeća pr. Kr. te je
ubrzo postala važan vojni i trgovinski centar. Rimska vojska
je od ovog perioda kontrolirala stanovništvo na obalnom
području sjevernog Jadrana i okončala ratne sukobe između
raznih ilirskih i istarskih plemena. Za vrijeme Cezara Terge-
sta i Pola ova regija dobila je status kolonije što je uvjetovalo
nagli gospodarski razvitak Istarskog poluotoka. Oktavijan
August upoznao je područje između Drave i Save za vrijeme
ratova oprotiv Japoda (35 - 31 pr. Kr.). Općenito se sma-
tra kako su August i njegov bliži krug suradnika pokazivali
interes za Istru. Nakon što su došli u posjed zemljišta, ubrzo
zatim anektirali su i cijeli poluotok. Rimljani su započeli sa
sadnjom maslina i vinograda na zapadnim obalama Istre.
Između Pole i Tergeste otkriveni su tragovi nekoliko vila sa
postrojenjima za proizvodnju maslinovog ulja, no njihovi
vlasinci do sada nisu identificirani. Ipak, poznate su dvije
radionice amfora na ovom području. Jedna je locirana u
blizini Poreča (Parentium) u uvali Loron. Druga radionica
nalazila se južno od Fažane, preko puto Brijunskog otočja.
Za vrijeme prvog perioda vlasnici radionice bili su članovi
ugledne i utjecajne rimske obitelji Laecani. Na Brijunima su
identificirane tri vile (Verige – Val Catena, Colci – Monte
Collisi, Castrum). Žigovi na amforama govore u prilog
BEZECZKY_Text.indd 205
205
činjenici kako su vile također bile u posjedu obitelji Laecani.
Veliki broj amfora pruža dobru osnovu za istraživanje trgo-
vinskih veza vile Castrum.
PRIJE POLOVICE PRVOG STOLJEĆA PR. KR.
Na temeljima vile Castrum nalazi se donekle misteriozna
građevina iz vremena kasne republike sa nekoliko nalaza,
uključujući i amfore tipa Lamboglia 2. Ove jadranske amfore
iz prve polovice prvog stoljeća pr. Kr. predstavljaju jednu od
najranijih indicija o uvezenom vinu iz Italije. Najranija faza
vile okončala je prije polovice prvog stoljeća pr. Kr.
POLOVICA PRVOG STOLJEĆA PR. KR. DO
POČETKA DRUGOG STOLJEĆA PR. KR.
Daleko veći broj amfora tipa Lamboglia 2 pripadao je drugoj
vili izgrađenoj u obliku slova U nakon polovice prvog sto-
ljeća pr. Kr. Neki od crijepova (tegulae) korištenih pri gradnji
druge vile potječu iz sjeverne Italije, uz prisutnu lokalnu pro-
izvodnju1. Druga vila izgrađena i korištena je od strane čla-
nova obitelji Laecani. Čitav dizajn građevine, podrum kao
i postrojenja za proizvodnju ulja (prese, lacus, dolia) iden-
tični su kao i kod druge dvije vile na Brijunima (Verige - Val
Catena, Kolci – Monte Collisi). Amfore i brojna keramika
(čepovi, pločice, spicae, cijevi za grijanje i lampe) izrađeni su
u radionici u Fažani preko puta Brijuna. U okviru rane faze
vile pronađene su i amfore tipa Dressel ante 6B bez žiga. Niti
nakon petrografske analize nije bilo moguće odrediti da li se
radi o importiranim primjercima ili pak o posudama veza-
nim uz lokalnu produkciju maslinovog ulja. Jedna amfora
tipa Dressel ante 6B sa žigom C·LAECANI·P·F pronađena
je međutim u blizini Venecije. Ovo je ujedno i dokaz kako
je obitelj Laecani proizvodila rane amfore tipa Dressel 6B
i poslije Cezarove vlasti te da su očigledno participirali u
1
Vidi prilog Bernia Milleta u ovom izdanju: PANSAE·VIBI,
A·FAESONI·AF, C·CEIONI·MAXI, C·FLAVI kao i lokalni
C.LAECANI.P.F. i LAEC.
18.03.19 14:40
 206 SAŽETAK

proizvodnji i trgovini maslinovog ulja. Postoji niz naznaka POLOVICA DRUGOG STOLJEĆA DO KONCA
koje upućuju kako je ulje iz Istre transportirano u sjevernu TREĆEG STOLJEĆA
Italiju, posebice u naselja u dolini rijeke Po. Najvažnije trgo-
Prestankom proizvodnje ulja u Istri u vremenu nakon Hadri-
vinske veze uspostavljene su sa Padovom i Akvilejom, no
ana uspostavile su se nove prilike. Karakteristične amfore
amfore obitelji Laecani, kao i kasniji primjerci iz Carskog
tipa Dressel 6B nestale su sa tradicionalnih tržišta nakon što
razdoblja, mogu se pronaći u gotovo svim gradovima sve do
je proizvodnja istarskih ulja svedena na minimum. Trgovci
Torina. Preko Akvileje istarske su vile, uključujući i one na
su popunili ovaj nedostatak uvozom ulja iz Betike u His-
Brijunima, snabdijevele nove provincije (Recija, Noricum i
paniji koje se transportiralo u amforama tipa Dressel 20.
Panonia) maslinovim uljem. Od druge polovice prvog sto-
Ovim uljem snabdijevani su svi centri u koje se prije toga
ljeća pr. Kr. ulje je proizvođeno i u sjevernoj Italiji te također
isporučivalo ulje iz Istre. Nedostatak ulja bio je u toj mjeri
transportirano amforama tipa Dressel 6B. Nakon sredine
alarmantan da je ulje iz Betike importirano čak i u samu vilu
prvog stoljeća produkcija ulja u sjevernoj Italiji značajano je
Castrum. Na to ukazuje amfora tipa Dressel 20 sa žigom iz
opala. Početak proizvodnje ulja u vili Dragonera mogao bi
ovog perioda (br. 40). Postoje mišljenja kako je problem sa
se stoga vjerojatno objasniti potrebom povećanja produk-
prestankom proizvodnje ulja u Istri usko povezan sa Makro-
cije u istarskim vilama. Ovo je ujedno i period u kojem su
manskim ratovima koji su izbili u vrijeme Marka Aurelija i
istočne granice Panonije pomjerene do Dunava što je uvje-
koji su onemogućili transport ulja u druge provincije. No
tovalo slanje većeg broja vojnika u tu provinciju. Zadnji član
sjeverne provincije koristile su ulje iz Betike znatno prije
obitelji Laecani preminuo je u godini 78 poslije Krista bez
ovih događanja. Razlog se prema tome ipak treba tražiti u
nasljednika. Radionica u Fažani prešla je u posjed cara Ves-
promijeni proizvodnje ulja u Istri. Amfore iz Betike mogu
paziana. Promjena vlasništva vidljiva je osobito kroz nove
se pronaći do konca trećeg stoljeća, iako ne u većem broju.
žigove. Carska administracija preuzela je radionice, vile, dje-
Naročito zanimljiva je amfora tipa Dressel 20 iz druge tre-
latnike i svo osoblje uključeno u produkciju. Žigovi ukazuju
ćine drugog stoljeća poslije Krista koja na sebi ima grafit
na činjenicu kako je produkcija ulja kontinuirano trajala do
(Asiatici / sec( ) x k(alendas) mar(tias) Alb( )).
vladavine Hadriana.
Afričke amfore pojavljuju se u vili između sredine drugog
Importirani nalazi iz druge faze vile uključuju veći broj
stoljeća i sredine trećeg stoljeća (Africana 1 A/B i Africana
amfora za vino (Dressel 6A, Dressel 2-4, Gauloise 2, Rho-
II A). Moguće je da su neki od ovih primjeraka povezani sa
dian, Knidian, Koan, Dressel 5, Cretan i Agora F65/66).
nedostatkom maslinovog ulja. Posude iz Afrike dospijele
Ne postoje dokazi o lokalnoj proizvodnji ribljih umaka za
su na prostor sjevernog Jadrana i mnogo ranije, posebice u
ovaj period, no pronađen je manji broj uvoza iz Hispanije
Akvileju a manji broj i u provincije Noricum i Panoniju.2
(amfore tipa Dressel 7-11, Beltran II A i II B te Dressel 14A).
Sa jedne strane mrazom i ledom uništena stabla masline
Postoje jasne razlike u smislu trgovine između vile Castrum i
trebalo je ponovno zasaditi a sa druge strane primjetno je
područja sjeverno od Akvileje. Dok je u vilu Castrum dopre-
kako je izvoz ribljih umaka započeo najvjerojatnije upravo
mana očigledno tek manja količina ribljeg umaka iz Hispa-
u ovo vrijeme. U drugoj polovici drugog stoljeća pojavljuju
nije, u trgovački centar Magdalensberg na sjeveru kao i u
se novi tipovi amfora (Fažana 1 i 2) što jasno odslikava pro-
Panoniju dospijevele su znatno veće količine ovog začina.
mjene u poljoprivrednoj proizvodnji. Ovo promjene morale
Vila je vjerojatno nekoliko puta rekonstruirana i renovi-
su se istovremeno odraziti i na proizvodnju keramičke radi-
rana. Na ovo upućuju prije svega crijepovi iz vremena Tibe-
onice u Fažani. Iz ovog razdoblja poznajemo ime samo jed-
riusa, Kaligule i Nera kao i drugi primjerci iz prvog stoljeća
nog proizvođača, iako je izrada amfora pokrenuta i u drugim
poslije Krista. Ne postoje nikakve informacije o međusobnoj
mjestima kao što je primjerice Loron. U Fažani je pronađen
povezanosti istarskih radionica kao ni o vezama koje su poje-
žig Marka Aurelija Iustusa a njegov oltar otkriven je na gro-
dine radionice mogle imati sa radionicama u sjevernoj Ita-
blju uz vilu Castrum. Žig se nalazio na amfori tipa Fažana 1.
liji. Crijepovi (tegulae) iz Lorona pronađeni u vili Castrum
nosili su žigove SISENNAE i IMP·AVG·GER.
Između ova dva žiga stoji vremenski period od oko 60 do
2
Amfore tipa Tripolitana II transportirane su iz Aquileje u Magda-
lensberg i Emonu. Amfore tipa Africana I i Schöne-Mau XL pro-
70 godina. Kako i zašto su ovi crijepovi završili u vili ostaje nađene su u blizini Dunava u mjestu Ovilava (Noricum). Amfore
nepoznato, jer za niti jedno od ponuđenih objašnjenja ne tipa Leptiminus 1 pronađene su u mjestu Iseum u Savari duž Jantar-
može se tvrditi da je točno. skog puta. Jedan od najpoznatijih nalaza u regiji sjevernog Jadrana
su amfore tipa Africana I iz potunulog broda pronađenog u mjestu
Grado.

BEZECZKY_Text.indd 206 18.03.19 14:40

 CHAPTER 13 – Sažetak (Bezeczky)
Zasada se ne može reći kada i kako dugo je ovaj tip amfore
bio izrađivan, no njegova pojava svakako je povezana sa
ponovnom proizvodnjom maslinovog ulja. Posude ovog tipa
dosada nisu pronađene van Istre.
Vino je na početku uvoženo iz Italije. Kasnije se na Bri-
junima počelo također proizvoditi vino, premda ne znamo
točno od kada. U kasnom rimskom razdoblju vino iz Istre
spominje samo Cassiodorus. No broj preša upućuje da je
ipak proizvođena značajna količina vina. U slučaju da je vino
bilo izvoženo postavlja se pitanje u kojem tipu posuda? Kako
ne raspolažemo sa preciznim kronološkim podacima za sada
je nemoguće točnije datirati amfore za vino tipa Folimpo-
poli. Njihovo podrijetlo također ostaje otvoreno i treba ga
detaljnije ispitati.
Nakon završetka drugog stoljeća pojavljuju se novi tipovi
amfora za vino iz Male Azije, prije svega iz regija Perga-
mon i Efez (Agora M 125, M 126, M 240) kao i iz nekih
drugih dijelova Egeje (Agora G 199, Kapitän I i Kapitän
II). Nekoliko primjeraka amfora iz istočnog Sredozemlja
pronađeno je i u podvodnim istraživanjima u mjestu Val
Catena.3 Orijentalne amfore dosta su česte na prostoru
sjevernog Jadrana. U sjevernim provincijama njihov broj
osjetno je manji, sa izuzetkom amfora tipa Kapitän II koje su
u većim količinama pronađene duž Panonskog limesa.
Od sredine druge polovice trećeg stoljeća uvoze se i
afrička usoljena riba i/ili vino iz tuniških gradova Nabeul i
Thaenae (amfore tipa Africana II C i II D). Isti tip pronađen
je i u Akvileji. Vino (?) je od kasnog trećeg do ranog petog
stoljeća uvoženo i u amforama tipa Beltran 68 iz radionice
Gaditan u provinciji Betika.
ČETVRTO STOLJEĆE DO POČETKA PETOG
STOLJEĆA
Tko je koristio vilu? Nakon vremena M. Aureliusa Iustisa ne
raspolažemo nikakvim informacijama vezanim sa ovo pita-
nje. Čini se izglednim kako su Istra i Brijuni dospijeli pod
direktnu vlast Milanskog vikara, slijedeći reforme koje je
pokrenuo Dioklecijan. U ovom periodu vjerojatno je pokre-
nuta i fullonica koja je predstavljala apsolutno novu funkciju
unutar vile. Ulje je također vjerojatno proizvođeno. U slu-
čaju da se ulje uvozilo iz istočnog Sredozemlja, vrlo je vjero-
jatno kako su se za transport koristile amfore tipa LR1 i LR2
ili pak tipa Ephesos 56.
3
Bloier 2012, Agora M 126, no. 1421, M 240 nos. 1506, 1570 and
1575.
BEZECZKY_Text.indd 207
207
Ne raspolažemo također ni sa informacijama glede
moguće lokalne proizvodnje vina u ovom periodu. Ne treba
iznenaditi niti podatak kako je u vilu dospijevalo vino iz
Afrike i to iz mjesta Nabeul (Africana III A, III B i varijanta
tipa Keay 1B). Africana III jedan je od napoznatijih tipova
na području sjevernog Jadrana. Uvoz ribljih umaka iz Luzi-
tanije također je započeo u ovom periodu u amforama tipa
Almagro 50 (Keay XVI, Keay XXII), Almagro 51C i Keay
XIX (Almagro 51A-B).
PRVA POLOVICA PETOG STOLJEĆA DO PRVE
POLOVICE ŠESTOG STOLJEĆA
U petom stoljeću došlo je do izgradnje zidova koji su šti-
tili vilu te samom naselju dali konačnu konturu. Crkva Sv.
Marije podignuta je oko stotinu metara od naselja. Izgrad-
njom crkve zasigurno se promijenila i funkcija samog nase-
lja. Proizvodi vile iz ovog razdoblja su nepoznati. No prema
Cassidorusu, Istra se na početku šestog stoljeća smatrala regi-
jom sa dobrim i kvalitetnim grožđem, vinom, uljem, ostri-
gama (kamenicama) i ribom a vino i ulje su čak izvoženi i
u Ravenu.4 Ako je Istra doista bila tako bogata, postavlja se
pitanje zašto je uopće bilo potrebno uvoziti prehrambene
proizvode iz dalekih područja?
Pronađena je veća količina amfora iz Ciclije, sa grčkog
kopna, te iz Thasosa i iz Levanta (LR 1, LR 2, LR 4) kao i
amfore iz regije oko Efeza (Agora M 307, LR 3 i Ephesus
56). Na Brijunima su pronađene i amfore iz drugih regija
(Kreta i Crno more).
Između prve polovice petog i prve polovice šestog stoljeća
u naselje je dospjela znatna količina afričkih amfora. Prona-
đeno je dosta proizvoda iz Nabuela: spatheia tip 1, Keay 3/5,
35A i B, 55, i 57. Otkriveno su i tipovi Keay 59/8B iz južne
Bizacene, zatim tipovi Keay 27/36 sa sjeverozapada kao i tip
Keay 62Q nepoznate regionalne provinencije.
„Paul Reynolds zamijetio je ovaj trend na cijelom Jadranu
od sredine petog stoljeća, objasnivši ga ‚glavnim brodskim
putem ka Raveni‘ odnosno mogućnošću gradova i naselja
duž ove rute da profitiraju od uvoza i izvoza.“5
4
Cassiod. var. 22: „Ono što je Kampanja za Rim to je Istra za Ravenu,
plodna provincija sa obiljem žita, vina i ulja ili bolje rečeno kuhinjski
ormar glavnog grada.“ 24: „Već smo prije izdali naredbu da bi Istra tre-
bala poslati vino i ulje, kojeg ima u izobilju godine, u kraljevsku rezi-
denciju u Raveni.“
5
Reynolds 2010, 99; vidi također: Auriemma – Quiri 2007, 34.
18.03.19 14:40
 208 SAŽETAK

SREDINA ŠESTOG STOLJEĆA DO KONCA SED- Naselje Castrum moralo je imati osobito bitnu ulogu u
MOG STOLJEĆA vremenu između petog i sedmog stoljeća. U šestom stoljeću
podignuta je i crkva Sv. Petra na oko tri stotine metara od
U ovom periodu afrički importi stizali su iz obalnog
naselja. Trenutno se ne može zaključiti da li je Castrum na
dijela centralne Bizacene u amforama tipa Keay 62 i 61 iz
Brijunima postao bizantijska vojna utvrda ili pak naselje za
Moknine, amforama tipa Keay 61C iz zaleđa Salakte, tipa
civilno pučanstvo. Činjenica kako se dvije spomenute crkve
Kaey 8A iz nepoznatog mjesta u Tunisu u predjelu Sahela
nalaze van zidina govori u prilog drugoj pretpostavci. Ovdje
kao i minijaturnim spatheia posudama tipa 3C iz Nabuela.
prezentirani dokazi jasno ukazuju kako se na Brijunima
Nešto je manji broj amfora iz istočnog Sredozemlja nego u
nedvojbeno nalazilo i Biskupska stolica koja je postojala više
razdoblju prije, radilo se vjerojatno samo o amforama tipa
od dva i pol stoljeća.
Ephesus LR 3, Ephesus 56, Samos Cistern type i amfora tipa
LRA 13 izrađena na otoku Kos. Zadnje navedeni primjerak
imao je i carski žig.

BEZECZKY_Text.indd 208 18.03.19 14:40

 APPENDIX

THE AMPHORA STAMPS OF THE FAŽANA

WORKSHOP

BEZECZKY_Text.indd 209 18.03.19 14:40

 1
210 APPENDIX

1 Late republican period: ca. 40 - ca. 25 B.C.

1.1 C·LAECANI·P·F

2.1 Augustan period (Magdalensberg I): ca. 25 - 10/5 B.C.

2.1 C·LAEC·A

2.2 Augustan period (Magdalensberg I): 10/5 B.C. - 14 A.D.

2.2 C·LAEC·BASSI FELIX·SER

3 Tiberian - early Claudian period (Magdalensberg II): 14 - 45/40 A.D.

3.1 LAEK·A

3.2 C· LAEK ADEL

3.3 C·LAEK ? ARCI

C·LAEK?
3.4 C·LAEK BAR

3.5 C·LAEC·B BARB

3.6 C·LAE CAESI

BEZECZKY_Text.indd 210 18.03.19 14:40

 2
The Fažana amphora stamps (Bezeczky) 211

3.7 C·LAEK CAR

3.8 C·LAEK·BAS CLARVS

3.9a LAE COM

3.9b CLAEKB COM

3.10 LAEK COMI

3.11 LAEK COMVS

·
3.12a C·LAEK·BAS EVCHAR

3.12b C·LAEK·BAS EVCHAR

3.13 LAE FA

3.14 C·LAEK FAV

BEZECZKY_Text.indd 211 18.03.19 14:40

 3
212 APPENDIX

3. 15 C·LAEK FELIX

3.16a C·LAEK FELIX·PET

3.16b C·LAEK·B FELIX·PETIL

3.17 C·LAEK·BASSI FELIX·SCR

3.18 C·LAEK FEL·TVR

3.19 LAE FVI

3.20a LAEK·H

3.20b LAEK H

3.21a LAEK HER

3.21b C·LAEK·B HER

BEZECZKY_Text.indd 212 18.03.19 14:40

 4
The Fažana amphora stamps (Bezeczky) 213

3.22a C·LAEK·BASS HERME

3.22b HERME HERME

3.23 LAEK HOM

3.24a C·LAEK·B IALIS

3.24b C·LAEK·B IALI

3.25 LAEK L

3.26 C·LAEK OPI

3.27 LAEK OPTA

3.28a C·LAEK SPERAT

3.28b C·LAEK·BAS SPERATVS

3.29 C·LAEK SYNT

BEZECZKY_Text.indd 213 18.03.19 14:40

 5
214 APPENDIX

3.30 C·LAEK VRBA

3.31a LAEK VIAT

3.31b LAEK VIAT

4 Post Magdalensberg period: 45/50 - 78 A.D.

4.1 C·LAEK·BASS AMETHYSTI

4.2 C·LAEK AMYCVS

4.3 C·LAEK·B BAR

4.4 C·L·B COLTO

4.5 C·L·B …ES

…ES
4.5 C·L·B C·L·B

4.6 C·LAEK·B CRESCENTIS

4.7a LAEK DI

BEZECZKY_Text.indd 214 18.03.19 14:40

 6
The Fažana amphora stamps (Bezeczky) 215

4.7b LAEK DI

4.8a EVCHAR

4.8b C·LAEK·BASSI EVCHARISTI

4.9 C·LAEK·B FAL

4.10 C·LAEK·B MARTI

4.11 [C·LAEK]·BAS NICOMEDE

4.12 C·LAEK·B PIERI

4.13 LAEK·B

4a Post Magdalensberg period: 78 A.D.

4a.1 C·LAEK·BAS CLYMEN

4a.2 [C]·LAEK·B DATI

4a.3 C·LAEK·BAS PAGANI

BEZECZKY_Text.indd 215 18.03.19 14:40

 216 APPENDIX

4a.4 C·LAEK·BAS PTOLEM

5 Vespasian period: 78 - 24th June 79 A.D.

5.1 IMPE·VESP BARNAE

5.2a IMP CLYME

5.2b IMP CLYME

5.2c IMP·CAES·VESP·AVG CL[YME]

5.3 IMP·V[ESP] COLONI

5.4a IMP DAT

5.4b IMP·CAES·VESP DAT

5.5a IMP PAGANI

5.5b IMPE·VESP PAGANI

5.6 IMP POLL

5.7 IMP·CAES·VESP PTOLEM

BEZECZKY_Text.indd 216 18.03.19 14:40

 The Fažana amphora stamps (Bezeczky) 217

5.8 IMPE·VESP

5.9 IMP …TI and …SI

5.10 IMP LESBI

6 Titus period: 25th June 79 - 13th September 81 A.D.

6.1a IMP·T BERENT

6.1b IMP·T· CAES BERENTS

6.2 IMP·T· CAE·AVG PRIMIGEN

7 Domitian period: 14th September 81 - 18th September 96 A.D.

7.1 IMP·DOMITI LESBI

7.2 IMP·DOM LESBI

7.3 IMP·AVG SPENDO

8 Nerva period: 19th September 96 - 27th January 98 A.D.

8.1 IMP·NERVE

BEZECZKY_Text.indd 217 18.03.19 14:40

 9
218 APPENDIX

9 Trajan period: 28th January 98 - 9th August 117 A.D.

9.1 IMP·TRA MANI

9.2 IMP·TRA SERV...

9.3 IMP·TRA ...VS

10 Hadrian period: 10th August 117 - 10th July 138 A.D.

10.1 HAD·AVG

11 M. Aurelius Iustus: c. 160 A.D. -

11.1 M AVRELI/ IVSTI

11.2 F A

12 Period?: without dominus stamp

12.1 EPAPHRO 12.3 PIERI

12.2 FVS 12.4 ROM

BEZECZKY_Text.indd 218 18.03.19 14:40

 219

PL ATE INDEX

Amphorae 31 11d, 12a1, 12a2, 12a3, 12a4, 12a5, 12a6, 12a7

32 12a8, 12a9, 12a10, 12a11, 12a12, 12b, 12c1, 12c2, 12d1, 12d2
Plates Catalogue nos. 33 12e, 12f, 12g, 12h, 12i, 13a, 13b1, 13b2, 14a1, 14a2, 14a3,
1 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 14a4
2 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24 34 14a5, 14a6, 14a7, 14b1, 14b2, 14b3, 15, 16, 17a1, 17a2, 17a3,
3 25, 26, 27, 28, 29, 30, 31, 32, 33 18, 19
4 34, 35, 36, 37, 38, 39, 40, 42, 43, 44, 45, 46, 47, 48
5 41, 49, 50, 51 Photomicrographs of the raw materials and amphorae
6 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64
7 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78 Plates
8 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91 35 1-8
9 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104 36 1-8
10 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 37 1-8
117, 118
11 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130,
131
12 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143,
144, 145, 146, 147
13 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159,
160, 161, 162, 163
14 164, 165, 166, 167, 168, 169, 170, 171, 173, 174, 175, 176,
177, 178
15 172, 179, 180, 181, 182, 183, 184
16 2, 3, 5, 7, 11, 12, 13, 14, 15, 16, 17, 18, 19
17 20, 21, 22, 38, 40, 86, 100, 108, 123

Photomicrographs of the amphorae

Plates Catalogue nos.

18 9, 11, 12, 13, 14, 15, 16, 17
19 18, 19, 20, 21, 22, 50, 51, 108
20 110, 111, 117, 128, 131, 132, 133, 134
21 135, 136, 137, 140, 141, 143, 146, 147
22 150, 151, 153, 160, 164, 165, 166, 167

Photos of the tile stamps

Plates Catalogue nos.

23 1, 2, 3, 4a1, 4a2, 4a3, 4a4, 4a5, 5a1, 5a2, 6a1
24 6a2, 6a3, 6a4, 6a5, 6a6, 7a1, 7a2, 7a3, 7a4
25 7b1, 7b2, 7b3, 7b4, 7b5, 7b6
26 7b7, 7b8, 7b9, 7b10, 7b11, 7b12, 7b13, 7b14, 7b15, 7b16
27 7b17, 7b18, 7b19, 7b20, 7b21, 7b22, 7b23-Gnirs 1908,
7b23-Matijašić 1987, 7b24, 7b25
28 7b26, 7b27, 7b28, 7b29, 7b30, 7b31, 7b32, 7b33, 7b34, 7b35,
7b36, 7b37
29 7b38, 7b39, 7b40, 8a1, 8b1, 8b2, 8b3, 8b4, 8c, 9a1, 9a2-
Gnirs 1908, 9a2-Matijašić 1987, 9a3, 9a4, 9a5, 9a6, 9a7,
10
30 11a, 11b1, 11b2, 11b3-Gnirs 1904/1910, 11b3-Matijašić
1987, 11b4, 11b5, 11b6, 11b7, 11b8, 11c1, 11c2, 11c3, 11c4

BEZECZKY_Text.indd 219 18.03.19 14:40

BEZECZKY_Text.indd 220 18.03.19 14:40
 A MPHOR AE T YPE INDEX

ITALIAN and ADRIATIC Koan, nos. 63, 64, 65

Dressel 5, no. 66
Lamboglia 2, nos. 1, 2, 3 Crétoise AC 4, nos. 67, 68, 69
Dressel 6A, nos. 4, 5, 6, 7, 8, Agora G 199 / Pinched handle, no. 70
Dressel 6B earliest form (1A phase) / ante Dressel 6B, nos. 9, 10 Agora F 66, no. 71
Dressel 6B, nos. 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25 Agora M 125, no. 72
Fažana 1, no. 26 Agora M 279 similis, no. 73
Fažana 2, no. 27 Agora M 126, nos. 74, 75, 76
Porto Recanati, nos. 28, 29 Agora M 240, no. 77
Forlimpopoli, nos. 30, 31 35 Kapitän 1, no. 78
Aquincum 78, nos. 32, 33 Kapitän 2, no. 79
Dressel 2-4, no. 34, 35, 36 Agora M 273, no. 80
Late Roman Amphora 1 (LR 1), nos. 81, 82, 83, 84, 85, 86
SPANISH Late Roman Amphora 2 (LR 2), nos. 87, 88
Late Roman Amphora 3 (LR 3), nos. 89, 90
Dressel 7-11, no. 37 Ephesus 56 / Ephesus 8, nos. 91, 92
Beltran IIA, no. 38 Agora M 307, nos. 93, 94, 95, 96
Dressel 14A, no. 39 Late Roman Amphora 4 (LR 4) / Gaza, nos. 97, 98, 99
Dressel 20, nos. 40, 41, 42, 43, 44, 45 Late Roman Amphora 13 (LR 13), no. 100
Almagro 50A/ Keay XXII, no. 46 Samos Cistern Type, no. 101
Almagro 51A-B, nos. 47, 48 Beirut, no. 102
Almagro 51C, no. 49 Cretan: AC 1C/D (Marangou 1994), MRC 3 (Portale – Romero 2000),
Beltran 68, nos. 50, 51 nos. 103, 104
Égyptienne Bitroconique 3, no. 105
GAULISH
BLACK SEA
Gauloise 4, no. 52
Sinope C II type / Kuzmanov 9?, nos. 106, 107
SICILIAN
AFRICAN
Mid Roman 1, nos. 53, 54
Hammamet 3A, no. 108
ITALIAN or EASTERN MEDITERRANEAN Tripolitanian III?, no. 109
Africana I A/B, no. 110
Palatine East 1/LRA 1, nos. 55, 56 Africana I variant?, no. 111
Africana II A?, no. 112
EASTERN MEDITERRANEAN Africana II B, Pseudo-Tripolitanian?, no. 113
Africana II C3, no. 114
Crete TRC 2 (Portale-Romero 2000), no. 57 Africana II C, no. 115
Rhodian, nos. 58, 59, 60 Africana II D, no. 116
Knidian, no. 61 African II/III transitional, no. 117
Koan type with pinched handle, no. 62 Africana II variant, nos. 118, 119

BEZECZKY_Text.indd 221 18.03.19 14:40

 Africana III A, nos. 120, 121, 122 MISCELLANEOUS
Africana III A/B, no. 123
Africana III B, nos. 124, 125, 126 Sinope C-I or C-II, no. 170
Africana III C, no. 127 Keay 22, small size Almagro 50, no. 171
Spatheion 1 ? (or Keay 35A), nos. 128, 129, 130 Gauloise 2, no. 172
Spatheion 1 late, no. 131 Torone VII, no. 173
Spatheion 3A, no. 132 Torone VII / Adamsheck 22, no. 174
Spatheion 3B, no. 133 Asia Minor production?, no. 175
Spatheion 3C, nos. 134, 135, 136 Late LRA 1, no. 176
Keay 3/5 or 64, no. 137 Crypta Balbi II, no. 177
Keay 8B, no. 138 TRC 2, no. 178
Keay 59 and Keay 8B / Bonifay 37-38-39, no. 139 TRC 1 or 2, no. 179
Keay 11B var. Keay 1984, Fig. 172.2, no. 140 TRC 4, no. 180
Keay 35A, no. 141 Miscellaneous, no. 181
Keay 35B, no. 142 Miscellaneous, no. 182
Keay 36, no. 143 LR 3 or Ephesus 56, no. 183
Keay 55, nos. 144, 145 Miscellaneous, no. 184
Keay 57, nos. 146, 147, 148
Keay 57-56-55, no 149
Keay 62 Q or Albenga 11/12, nos. 150, 151, 152
Keay 62, nos. 153, 154, (Keay 62 or 61?) no. 155
Keay 61, nos. 156, 157
Keay 61A/D, nos. 158, 159
Keay 61 ?, no. 160
Keay 61C, no. 161
Keay 8A / Bonifay 50, nos. 162, 163
Keay 34, nos. 164, 165, 166
Keay 1B?, no. 167
Storage jars?, nos. 168, 169

BEZECZKY_Text.indd 222 18.03.19 14:40


BIBLIOGR APHY
AAD (1977) Aquileia e l’Oriente Mediterraneo. Settimana di studi aqui-
leiesi 7, Antichità Altoadriatiche 12, Udine.
Abad Casal, L. (1975) El Guadalquivir, vía fluvial romana, Sevilla.
Adamsheck, B. (1979) Kenchreai. Eastern Port of Corinth. IV The Pot-
tery, Leiden.
Aldini, T. (1978) Anfore forolimpopilensi, Archeologia Classica, 230-
245.
Aldini, T. (1981) Fornaci di Forum Popili, Forlimpopoli.
Anselmino, L. – Coletti, C. M. – Ferrantini, M. L. – Panella,
C. (1986) Ostia. Terme del Nuotatore, In A. Giardina (ed), Società
romana e impero tardoantico III. Le merci, gli insediamenti, Roma,
45-81.
Arthur, P. (1990) Anfore dall’alto Adriatico e il problema del Samos
Cistern Type, Aquileia Nostra 61, 281-296.
Arthur, P. (1998) Eastern Mediterranean amphorae between 500 and
700: a view from Italy, In L. Sagui (ed), Ceramica in Italia: VI-VII
secolo, Firenze, 157-184.
Augenti, A. – Cirelli, E. (2010) Classe: un osservatorio privilegiato
per il commercio della Tarda Antichità, In S. Menchelli – S. Santoro –
M. Pasquinucci – G. Guiducci (eds), LRCW 3. Late Roman Coarse
Wares, Cooking Wares and Amphorae in the Mediterranean, BAR Int.
series 2185, Oxford, 605-615.
Aulas, C. (1980-81) Marques et inscriptions sur amphores trouvées à
Roanne, Cahiers Archeologiques de la Loire 1, 61-91.
Auriemma, R. (2000) Le anfore del relitto di grado e il loro contenuto,
MEFRA 112-1, 27-51.
Auriemma, R. (2006) Relitti lungo la costa occidentale dell’Adriatico,
In S. Čače – A. Kurilić – F. Tassaux (eds), Les roues de l’Adriatique
antique. Géographie et économie / Putovi antičkog Jadrana. Geografija i
gospodarstvo, Bordeaux – Zadar, 167-177.
Auriemma, R. (2007) Anfore, In C. Morselli (ed), Trieste Antica. Lo
scavo di Crosada, Trieste, 133-154.
Auriemma, R. – Degrassi, V. – Gaddi, D. – Mauro, S. – Oriolo,
F. – Riccobono, D. (2007) Strutture portuali minori in area alto­
adriatica. Il Progetto Interreg Italia-Slovenija “AltoAdriatico”, AAAd
65, 497-537.
Auriemma, R. – Degrassi, V. – Mauro, P. – Oriolo, F. – Ricco-
bono, D. (2008) Terre di mare: paesaggi costieri dal Timavo alla peni-
sola muggesana, In R. Auriemma – S. Karinja (eds), Terre di Mare, Atti
del Convegno Internazionale di Studi, Trieste, 8-10 novembre 2007,
Trieste – Pirano, 75-211.
Auriemma, R. – Degrassi, V. – Quiri, E. (2012) Produzione e circo-
lazione di anfore in Adriatico tra III e IV secolo. Nuovi dati da conte-
sti emblematici, In Ceramica romana nella Puglia Antica, seminario di
studio 2012, Bari, 153-195.
Auriemma, R. – Degrassi, V. – Quiri, E. (2015) Eastern amphora
imports in the Adriatic Sea: evidence from terrestrial and underwater
contexts of the Roman Imperial age, In S. Demesticha (ed), Per
Terram, Per Mare. Seaborne Trade and the Distribution of Roman
Amphorae in the Mediterranean, Uppsala, 139-160.
BEZECZKY_Text.indd 223
223
Auriemma, R. – Quiri, E. (2004) Importazioni di anfore orientali
nell’Adriatico tra primo e medio impero, In Transport Amphorae and
Trade in the Eastern Mediterranean. Acts of the International Collo-
quium at the Danish Institute at Athens, Aarhus, 43-55.
Auriemma, R. – Quiri, E. (2007) La circolazione delle anfore in Adri-
atico tra V e VIII sec. d.C., In S. Gelichi – C. Negrelli (eds), La circo-
lazione delle ceramiche nell’Adriatico tra tarda antichità e alto medioevo.
III Incontro di studio CER.AM.IS, Documenti di Archeologia 43, Man-
tua, 31-64.
Auriemma, R. – Degrassi, V. – Gaddi, D. – Maggi, P. (2016)
Canale. Anfora: uno spaccato sulle importazioni di alimenti ad Aqui-
leia tra I e III secolo d.C., Antichità Altoadriatiche 84, 379-404.
Autret, C. – Yağci, R. – Rauh, N. K. (2010) Soli/Pompeiopolis’te
LRA 1 Amphora Firini Alani / An LRA 1 Amphora Kiln site at Soli/
Pompeiopolis, ANMED, Anadolu Akdenizi Arkeoloji Haberleri /
News of Archaeology from Anatolia’s Mediterranean Areas 8, 203-207.
Bakker, L. – Galsterer-Kröll, B. (1975) Graffiti auf römischer
Keramik im Rheinischen Landesmuseum Bonn, Epigraphische Studien
19, Bonn.
Baldacci, P. (1967-68) Alcuni aspetti dei commerci nei territori cisal-
pini, In Atti del Centro Studi e Documentazione sull’Italia Romana I,
Milano – Varese, 7-50.
Baldacci, P. (1972) Le principali correnti del commercio di anfore
romane nella Cisapina, In I problemi della ceramica romana di
Ravenna, della Valle padana e dell’alto Adriatico. Atti del Convegno
Internazionale, Ravenna, 10-12 Maggio 1969, Bologna, 103-131.
Bandelli, G. (1983) Per una storia della classe dirigente di Aquileia
romana, In Les “Bourgeoisies” municipales italiennes aux IIe et Ier siècles
av. J.-C., Atti del Convegno Internazionale Napoli, 7-10 dicembre 1981,
Roma, 175-203.
Begović Dvoržak, V. (1990) Anticka vila u uvali Verige na Brijunima,
VAMZ 23, 97-110.
Begović Dvoržak, V. (2001) Fortifikacioni sklop Kasrum – Petrovac
na Brijunima / Late antique and early medieval fortification complex
castrum – Petrovac on Brijuni Islands, Histria Antiqua 7, 177-190.
Begović Dvoržak, V. – Schrunk, I. (2006) Pottery and other finds
from the Maritime Villa in Verige, Brijuni / Keramika i drugi nalazi
maritimne vile u uvali Verige na Brijunima, Histria Antiqua 14, 191-
202.
Begović, V. – Schrunk, I. (2007a) The Brijoni Islands. Past, Architec-
ture and Cultural Heritage, Zagreb.
Begović Dvoržak, V. – Schrunk, I. (2007b) Production of wine and
olive oil in the Early Empire on the Brijuni Islands, Histria Antiqua
15, 327-340.
Begović, V. – Schrunk, I. (2009) Tranformacija rimskih vila na Brijun-
skom otočju / Transformations of Roman villas on the Brijuni Island,
Histria Antiqua 18, 223-236.
Begović Dvoržak, V. – Schrunk, I. (2010) Villa rustica u uvali
Dobrika (Madona) no otočju Brijuni / The villa rustica in Dobrika
(Madona) Bay on the Brijuni Islands, PriloziZagreb 27, 249-266.
18.03.19 14:40
 224 BIBLIOGRAPHY
Begović Dvoržak, V. – Schrunk, I. (2011a) A Late Antique Settle-
ment in Madona Bay, Brijuni Islands, Histria Antiqua 20, 375-390.
Begović Dvoržak, V. – Schrunk, I. (2011b) Osnivanje novih bisk-
upija u VI. Stoljeću na primjeru Brijuna, In Zbornik u čast Emilija
Marina za 60. rođenan, Split, 657-672.
Begović Dvoržak, V. – Schrunk, I. – Tutek, I. (2007) Crkva Sv.
Marije, građena uz castellum u uvali Madona, Brijuni, Kasnoantičko
i bizantsko razdoblje / The Church of St. Mary nearby castellum in
Madona Bay, Brijuni. Late Roman and Byzantine period, PriloziZa-
greb 24, 229-240.
Belotti, C. (2004) Ritrovamenti di anfore romane a Iulia Concordia,
Gruaro.
Belotti, C. (2008a) Anfore di Sinope in Italia settentrionale, ReiCret­
Acta 40, 279-284.
Belotti, C. (2008b) Merci, transporti e contatti nella Venetia: le impor-
tazioni di anfore olearie dal Mediterraneo orientale, In R. Auriemma –
S. Karinja (eds), Terre di Mare. L’archeologia dei paesaggi costieri e le
variazioni climatiche, Trieste, 454-461.
Beltran Lloris, M. (1970) Las Ánforas Romanas en España, Zaragoza.
Bentaher, F. (1994) General account of recent discoveries at Tocra,
Lybian Studies 25, 231-243.
Bernal, D. (2009) Ánforas y vino en la Antigüedad Tardía. El ejemplo de
la Hispania meridional, In J. Blánquez Pérez – S. Celestino Pérez (eds),
El vino en época tardoantigua y medieval, Madrid, 33-60.
Bernal, D. (2016a) Beltrán 68 (Costa de Baetica), In Amphora ex His-
pania. Paisajes de producción y de consumo, http://amphorae.icac.cat/
tipol/view/29.
Bernal, D. (2016b) Beltrán 72 (Costa de Baetica), In Amphora ex His-
pania. Paisajes de producción y de consumo, http://amphorae.icac.cat/
amphora/beltran-72-baetica-coast.
Bernal, D. (2016c) Lusitanian Amphorae in the Strait of Gibraltar:
Interprovincial Food Supply, In I. Vaz Pinto – R. R. De Almeida –
A. Martin (eds), Lusitanian Amphorae: Production and Distribution,
Roman and Late Antique Mediterranean Pottery 10, Oxford, 299-310.
Bernal Casasola, D. (1996) Las ánforas del tipo Beltrán 68 en Hispa-
nia: problemática y estado de la cuestión, In Actas del XXIII Congreso
Nacional de Arqueología II, Elche, 251-269.
Bernal Casasola, D. (1998) La producción anfórica en la Bahía de
Algeciras en época romana, In D. Bernal (ed), Excavaciones Arqueo-
lógicas en el alfar romano de Venta del Carmen (Los Barrios, Cádiz),
Madrid, 19-42.
Bernal Casasola, D. (2000a) La producción de ánforas en la Bética
en el s. III y durante el Bajo Imperio, In D. Bernal – E. García Vargas
(eds), Ex Baetica Amphorae. Conservas, aceite y vino de la Bética en el
Imperio Romano I, Écija, 239-372.
Bernal Casasola, D. (2000b) Las ánforas béticas en los confines del
Imperio. Primera aproximación a las exportaciones en la pars orientalis,
In D. Bernal – E. García Vargas (eds), Ex Baetica Amphorae. Conservas,
aceite y vino de la Bética en el Imperio Romano III, Écija, 925-988.
Bernal Casasola, D. – García Vargas, E. (2008) Ánforas de la
Bética, In D. Bernal – A. Ribera i Lacomba (eds), Cerámicas hispanor-
romanas. Un estado de la cuestión, Cadiz, 661-687.
Berni Millet, P. (1998) Las ánforas de aceite de la Bética y su presencia
en la Cataluña romana, Col∙lecció Instrumenta 4, Barcelona.
Berni Millet, P. (2008) Epigrafía anfórica de la Bética, nuevas formas
de análisis, Col·lecció Instrumenta 29, Barcelona.
Berni Millet, P. – Gorostidi Pí, D. (2013) C. Iulius Valerianus et
C. Iulius Iulianus. Mercatores del aceite bético en un signaculum de
plomo para ánforas Dressel 20, JRA 26, 113-135.
Berni Millet, P. – Moros Díaz, J. (2012) Los sellos in planta pedis
de las ánforas olearias béticas Dressel 23 (primera mitad siglo V d.C.),
Archivo Español de Arqueología 85, 193-219.
BEZECZKY_Text.indd 224
Bertacchi, L. (1992) Il culto delle Dóminae ad Aquileia. Traccia per
una ricerca storico-topografica, Aquileia Nostra 63, 9-50.
Berti, F. (1990) Fortuna maris: la nave romana di Comacchio, Bologna.
Bertucci, G. (1991) Les amphores et le vin de Marseille: 6e. s. avant
J.-C. -2e. s. après J.-C., Révue Archéologique de Narbonnaise Suppl. 25,
Marseille.
Bezeczky, T. (1985) P. Iturius S. amphorája Vetus Salinában (P. Iturius
Sabinus’ amphora in Vetus Salina), Folia Archaeologica 36, 69-74.
Bezeczky, T. (1987) Roman amphorae from the Amber Route in West-
ern Pannonia, BAR Int. series 386, Oxford.
Bezeczky, T. (1993) Gari Flos Hispanici, In Ptujski Arheološki Zbornik,
Ptuj, 241-250.
Bezeczky, T. (1994) Amphorenfunde vom Magdalensberg und aus Pan-
nonien. Ein Vergleich, Kärntner Museumschriften 74, Klagenfurt.
Bezeczky, T. (1995) Roman Amphora Trade in Pannonia, In La Panno-
nia e l’Impero Romano, Atti del convegno internazionale, Annuario
dell’Accademia d’Ungheria, Roma – Milano, 155-175.
Bezeczky, T. (1997) Amphorae from the auxiliary fort of Carnuntum,
In H. Stiglitz (ed), Das Auxiliarkastell Carnuntum 1, Forschungen
1977-1988, SoSchrÖAI 29, Wien, 147-178.
Bezeczky, T. (1998a) The Laecanius Amphorae Stamps and the Villas
of Brijuni, DenkschrWien 261, Wien.
Bezeczky, T. (1998b) Amphora types of Magdalensberg, Arheološki
Vestnik 49, 225-242.
Bezeczky, T. (2000) Baetican amphorae from the north-eastern prov-
inces of the Roman Empire, In D. Bernal – E. García Vargas (eds), Ex
Baetica Amphorae. Conservas, aceite y vino de la Bética en el Imperio
Romano, Écija, 1359-1370.
Bezeczky, T. (2005a) Dressel 6A, Lamboglia 2, In S. Keay – D. F. Williams,
Roman Amphorae: a digital resource, University of Southampton, http://
archaeologydataservice.ac.uk/archives/view/amphora_ahrb_2005/.
Bezeczky, T. (2005b) Roman amphorae from Vindobona, In F. Krin-
zinger (ed), Vindobona. Beiträge zu ausgewählten Keramikgattungen in
ihrem topographischen Kontext, Wien, 35-83.
Bezeczky, T. (2006) Consumer behaviour in Ephesus in the first cen-
tury B.C., In D. Malfitana – J. Poblome – J. Lund (eds), Old Pottery
in a New Century. Innovating Perspectives on Roman Pottery Studies,
Catania, 287-308.
Bezeczky, T. (2008) Amphorae from the West. Evidence of the long dis-
tance trade connection with Ephesus, In M. Hainzmann – R. Wede-
ning (eds), Akten des 2. Internationalen Kolloquiums Klagenfurt, Ins-
trumenta Inscripta Latina 2, Klagenfurt, 25-34.
Bezeczky, T. (2010) Trade connections between Ephesus and Adriatic
region, Histria Antiqua 19, 351-358.
Bezeczky, T. (2013a) The Amphorae of Roman Ephesus, FiE XV/1,
Wien.
Bezeczky, T. (2013b) Cypriot Amphorae in Ephesus, In M. L. Lawall –
J. Lund (eds), The Transport Amphorae and Trade of Cyprus, A Semi-
nar at the Danish Institute at Athens, 23rd-24th of April 2007, Gösta
Enbom Monographs 3, Aarhus, 165-167.
Bezeczky, T. (2014) The Laecanius Amphorae, In Roman Pottery and
Glass Manufactures; Production and trade in the Adriatic region. Exper-
imental archaeology. Proceedings of the 2nd International Archaeologi-
cal Colloquium, 28-29 October 2011, Crikvenica, Croatia, Crikvenica,
241-257.
Bezeczky, T. – Berni Millet, P. – González Cesteros, H. (2015)
New research on the Castrum villa, on the Island of Brijuni (Croatia),
Preliminary report, In S. Demesticha (ed), Per Terram, Per Mare. Sea-
borne Trade and the Distribution of Roman Amphorae in the Mediter-
ranean, Uppsala, 189-198.
Bezeczky, T. – Mange, M. A. (2009) New petrographic data on the
late phase of the Laecanius workshop in Fažana (Istria, Croatia), In
18.03.19 14:40
 BIBLIOGRAPHY
Vessels: Inside and Outside, 9th European Meeting on Ancient Ceramics
(EMAC ’07), 24-27 October 2007 Budapest, Budapest, 247-254.
Bezeczky, T. – Pavletić, M. (1996) New objects from the figlina of
C. Laecanius Bassus, ÖJh 65, 143-163.
Blanc-Bijon, V. – Carre, M.-B. – Hesnard, A. – Tchernia, A.
(1998) Recueil de timbres sur amphores romaines (1987-1988), Aix-en-
Provence.
Bloier, M. (2012) Die taucharchäologischen Untersuchungen in der Val
Catena-Bucht auf Brioni (Istrien/Kroatien) 1996/97. Befunde und
Funde, Passauer Universitätsschriften zur Archäologie 14, Rahden.
Bombico, S. (2016) Lusitanian Amphorae on Western Mediterranean
Shipwrecks: Fragments of Economic History, In I. Vaz Pinto – R. R.
De Almeida – A. Martin (eds), Lusitanian Amphorae: Production
and Distribution, Roman and Late Antique Mediterranean Pottery 10,
Oxford, 445-460.
Bonifay, M. (2004) Études sur la céramique romaine tardive d’Afrique,
BAR Int. series 1301, Oxford.
Bonifay, M. (2013) Africa: patterns of consumption in coastal regions
vs. inland regions. The ceramic evidence (300-700 AD), In L. Lavan
(ed), Local Economies? Production and Exchange of Inland Regions
in Late Antiquity, Late Antique Archaeology 10, Leiden, 529-566.
Bonifay, M. – Botte, E. – Capelli, C. – Contino, A. – Djaoui, D.
– Panella, C. – Tchernia, A. (2015) Nouvelles hypothèses sur
l’origine et le contenu des amphores africaines Ostia LIX et XXIII,
Antiquités africaines 51, 189-210.
Bonifay, M. – Capelli, C. – Drine, A. – Ghalia, T. (2010) Les pro-
ductions d’amphores romaines sur le littoral tunisien: archéologie et
archéométrie, ReiCretActa 41, 319-327.
Bonifay, M. – Capelli, C. – Moliner, M. (2011) Les amphores afri­
caines de la basilique de la rue Malaval à Marseille (Ve siècle), In SFE-
CAG, Actes du congrès d’ Arles, Marseille, 235-254.
Bonifay, M. – Capelli, C. (2013) Les Thermes du Levant à Leptis
Magna: quatre contextes céramiques des IIIe et IVe siècles, Antiquités
africaines 49, 67-150.
Bonifay, M. – Cerova, Y. (2008) Importations de céramiques africaines
à Byllis, ReiCretActa 40, 37-44.
Bónis, É. (1942) A császárkori edénymuvesség termékei Pannoniában 1/
Die Kaiserzeitliche Keramik von Pannonien 1, Dissertationes Pan-
nonicae II, Budapest.
Bonnet, F. – Cattin, M. I. (1999) Le materiel archéologique: Cata-
logue systématique, In R. Kasser (ed), Explorations aux Qouçoûr
El-Izeila: lors des campagnes 1981, 1982, 1984, 1985, 1986, 1989 et
1990, Mission Suisse Arch. Copte Univ. Genève 3, 519. 573.
Borzić, I. (2011) Hispanski garum na Burnumskom stolu / Hispanic
garum at a Burnum table, Archaeologia Adriatica 5, 65-88.
Böttger, B. (1992) Die kaiserzeitlichen and spätantiken Amphoren aus
dem Kerameikos, AM 107, 315-381.
Bovini, G. (2006) Ravenna. Kunst und Geschichte, Ravenna.
Bratož, R. (1994) Der Einfluß Aquileias auf den Alpenraum und das
Alpenvorland (von den Anfängen bis um 700), In E. Boshof (ed), Das
Christentum im bairischen Raum. Von den Anfängen bis ins 11. Jahr-
hundert, Passauer historische Forschungen 8, Köln, 29-61.
Bratož, R. (1996) Christianisierung des Nordadria- und Westbalkan-
raums im 4. Jahrhundert, In R. Bratož (eds), Westillyricum und Nord-
ostitalien in der spätrömischen Zeit, Situla 34, Ljubljana, 299-366.
Bratož, R. (1998) Die römische Synode 680 und die Frage der Kir-
chenorganisation „in gentibus“ im 7. Jahrhundert, In Radovi
XIII. Međunarodnog kongresa za starokršćansku arheologiju / Acta
XIII. Congressus Internationalis Archaeologiae Christianae. Split - Poreč
25.9-1.10.1994, Split, 589-602.
Bratož, R. (2000) Das Patriarchat Grado im monotheletischen Streit,
In R. Bratož, Slovenija in sosednje dežele med antiko in karolinško
BEZECZKY_Text.indd 225
225
dobo. Začetki slovenské etnogeneze / Slowenien und die Nachbarlän-
der zwischen Antike und karolingischer Epoche, Anfänge der sloweni-
schen Ethnogenese 2, Situla 39, Ljubljana, 609-658.
Braun, J. (1924) Der christliche Altar in seiner geschichtlichen Entwick-
lung 1, München.
Brecciaroli Taborelli, L. (1987) Un contributo alla conoscenza
dell’impianto urbano di Eporeida (Ivrea): Lo scavo di un isolato a porta
Vercelli, Quaderni della Soprintendenza Archeologica del Piemonte 6,
97-157.
Brogiolo, G. P. (1999) S. Giulia di Brescia, gli scavi dal 1980 al 1992.
Reperti preromani, romani e alto medievali, Florence.
Brun, J.-P. (2004) Archéologie du vin et de l’huile dans l’Empire romain,
Paris.
Bruno, B. (1995) Aspetti di storia economica della Cisalpina Romana. Le
anfore di tipo Lamboglia 2 rinvenute in Lombardia, Roma.
Bruno, B. (1997) Contenitori da trasporto: i consumi di olio, vino e di
altre derrate, In F. Filippi (ed), Alba Pompeia, Archeologia della città
dalla fondazione alla tarda antichità, Alba, 516-532.
Bruno, B. (2002) Importazione e consumo di derrate nel tempio:
l’evidenza delle anfore, In F. Rossi (ed), Nuove ricerche sul Capitolium
di Brescia: scavi, studi e restauri, Milan, 277-308.
Brusić, Z. (1976) Byzantine amphorae (9th-12th century) from eastern
Adriatic underwater sites, Archaeologia Iugoslavica 17, 37-49.
Brusić, Z. (1999) Hellenistic and Roman relief pottery in Liburnia
(North-East Adriatic, Croatia), BAR Int. series 817, Oxford.
Brusić, Z. (2010) Early Medieval Finds from the Croatian Undersea
World, Archaeologia Adriatica 4, 243-255.
Buchi, E. (1973) Banchi di anfore romane a Verona. Note sui commerci
cisalpini, In Il territorio veronese in età romana. Atti del convegno tenuto
a Verona, 22-24 ott. 1971, Verona, 531-637.
Buchi, E. (1987) Asetto agrario, risorse e attività economiche, In
E. Buchi (ed), Il Veneto nell’età Romana I. Storiografia, organizzazione
del territorio, economia e religione I, Verona, 103-184.
Bueno, M. – Novello, M. – Mantovani, V. (2012) Progetto Aqui-
leia: Casa delle Bestie Ferite. Commercio e consumo ad Aquileia. Ana-
lisi delle anfore tardoantiche alla luce di alcuni contesti, ReiCretActa
42, 59-168.
Buiatti, A. M. (1990-91) L’agro sud-occidentale di Aquileia in età
romana (Ricerche sui materiali: i bolli laterizi), Tesi di laurea, Univer-
sità di Trieste.
Buiatti, A. M. (1994) Nuove acquisizioni sui bolli laterizi dell’agro
aquileiese, In Epigrafia della produzione e della distribuzione. Actes de
la VIIe Rencontre franco-italienne sur l’épigraphie du monde romain
(Rome, 5-6 juin 1992), CEFR 193, Rome, 415-431.
Bulić, D. (2009) Prilog poznavanju žigova na amforama fažanske radi-
onice / Contribution to knowledge about Stamps on Amphorae pro-
duced in the Fažana Workshop, Histria Antiqua 18-1, 257-270.
Bulić, D. – Koncani Uhač, I. (2009) Keramičarska radonica u Fažani,
resultati istraživanja 2007-2009 / Ceramic workshop in Fažana: The
results of investigations from 2007 to 2009, Histria Antiqua 17, 285-
298.
Bulić, D. – Koncani Uhač, I. (2011) Figlina u Fažani I njezina preo-
brazba u kasnoj antici / The pottery workshop at Fažana and its trans-
formation in the late Roman period, Histria Archeologica 41, 2010,
109-146.
Burnham, B. C. – Keppie, L. J. F. – Esmonde Cleary, A. S. – Tom-
lin, R.S.O. (1997) Roman Britain in 1996, Britannia 28, 460-472.
Buonopane, A. (2009) La produzione olearia e la lavorazione del pesce
lungo il medio e l’alto Adriatico: le fonti letterarie, In S. Pesavento
Mattioli – M.-B. Carre (eds), Olio e pesce in epoca romana. Produ-
zione e commercio nelle regioni dell’Alto Adriatico. Atti del convegno
(Padova, 16 febbraio 2007), Antenor Quaderni 15, Roma, 25-36.
18.03.19 14:40
 226 BIBLIOGRAPHY
Bustamante Álvarez, M. – Heras Mora, F. J. (2013) Producción
anfórica en Augusta Emerita (Mérida, Badajoz) y los nuevos hallazgos
del solar de la Escuela de Hostelería, In D. Bernal et al. (eds), Actas
del I Congreso Internacional de la Sociedad de Estudios de la Cerámica
Antigua en Hispania. Hornos, talleres y focos de producción alfarera en
Hispania, Cádiz, 331-345.
Cambi, N. (1983) Le anfore Dressel 20 nella Jugoslavia, In Producción y
comercio del aceite en la Antigüedad. Segundo Congreso Internacional,
Madrid, 363-389.
Cambi, N. (1989) Anfore romane in Dalmazia, In Amphores romaines et
histoire économique: dix ans de recherché, CEFR 114, 311-337.
Campanati, R. F. (2008) Temi musivi nei pavimenti d’area adriatica
(V-VI secolo). La cristianizzazione dell’Adriatico, Triest, 435-454.
Capelli, C. – Ben Lazreg, N. – Bonifay, M. (2006) Nuove pro-
spettive nelle ricerche archeometriche sulle ceramiche nordafricane:
l’esempio dell’atelier di Sullechtum-Salakta, Tunisia centrale (I-VI
secolo d.C.), In N. Cucuzza – M. Medri (eds), Archeologie. Studi
in onore di Tiziano Mannoni, Biblioteca di Archeologia 19, Florence,
291-294.
Capelli, C. – Bonifay, M. (2007) Archéométrie et archéologie des
céramiques africaines: une approche pluridisciplinaire, In M. Boni-
fay – J.-C. Tréglia (eds), LRCW 2. Late Roman Coarse Wares, Cook-
ing Wares and Amphorae in the Mediterranean. Archaeology and
Archaeometry, BAR Int. series 1662, Oxford, 551-567.
Capelli, C. – Bonifay, M. – Franco, C. – Huguet, C. – Leitch, V.
– Mukai, T. (2016), Étude archéologique et archéométrique intégrée,
In D. Malfitana – M. Bonifay (eds), La ceramica africana nella Sicilia
Romana / La Cèramique africaine dans la Sicile Romaine, Monografie
dell’IBAM 12, Catania, 273-352. 654-736.
Cargnelutti, C. (1992) Contributo per un catalogo dei bolli laterizi
aquileiesi, Ad Undecimum. Annuario 6, 98-111.
Cargnelutti, C. (1993) Contributo per un catalogo dei bolli laterizi
aquileiesi. Seconda parte, Ad Undecimum. Annuario 7, 82-102.
Carre, M.-B. (1985) Les amphores de la Cisalpine et de l’Adriatique au
debut de l’Empire, MEFRA 97, 207-245.
Carre, M.-B. (2007) L’évolution des importations à Aquilée III: les
amphores orientales: dones quantitatives compares, In G. Cuscito –
C. Zaccaria (eds), Aquileia dalle origini alla costituzione del ducato
longobardo: territorio, economia, società. Atti della 37 settimana
di studi aquileiesi, 18-20 maggio 2006, Antichità Altoadriatiche 65,
Trieste, 583-604.
Carre, M.-B. – Cipriano, M. T. (1985) Saggi di scavo a Seveliano. Le
anfore, Aquileia Nostra 56, 6-23.
Carre, M.-B. – Gaggadis-Robin, V. – Hesnard, A. – Tchernia,
A. (1995) Recueil de timbres sur amphores romaines 1987-1988, Aix-
en-Provence.
Carre, M.-B. – Maggi, P. – Merlatti, R. – Rousse, C. (2007) L’évo-
lution des importations à Aquilée: les nouvelles données de la fouille
au nord du Port Fluvial, Antichità Altoadriatiche 65, 539-632.
Carre, M.-B. – Pesavento Mattioli, S. – Belotti, C. (2009) Le
anfore da pesce Adriatiche, In S. Pesavento Mattioli – M.-B. Carre
(eds), Olio e pesce in epoca romana. Produzione e commercio nelle
regioni dell’Alto Adriatico. Atti del convegno (Padova, 16 febbraio
2007), Antenor Quaderni 15, Roma, 215-238.
Carre, M.-B. – Monsieur, P. – Pesavento Mattioli, S. (2014)
Transport amphorae Lamboglia 2 and Dressel 6A: Italy and/or Dal-
matia? Some clarifications, JRA 27, 417-428.
Carre, M.-B. – Maselli Scotti, F. (2001) Il porto di Aquileia: dati
antichi e ritrovamenti recenti, In C. Zaccaria (ed), Strutture portuali
e rotte marittime nell’Adriatico di età romana, CEFR 280, Trieste –
Rome, 211-243.
BEZECZKY_Text.indd 226
Carre, M.-B. – Pesavento Mattioli, S. (2003) Tentativo di clas-
sificazione delle anfore olearie adriatiche, Aquileia Nostra 74, 453-
475.
Carreras Monfort, C. (1999) Miscelánea: Las otras ánforas del
Monte Testaccio, In J. M. Blázquez Martínez – J. Remesal Rodríguez
(eds), Estudios sobre el Monte Testaccio (Roma) I, Col·lecció Instru-
menta 6, Barcelona, 91-98.
Casalini, M. – Crespi, M. (2010) Anfore tardoantiche di piccole
dimensioni a fondo piatto dalle pendici nord-orientali del Palatino.
Nuovi dati alla luce di un riesame tipologico e petrografico, ReiCret­
Acta 41, 101-111.
Ceazzi, A. – Del Brusco, A. (2014) La ceramica comune, la ceramica
da cucina locale e importata e le anfore dello scavo di via Bolovia,
Aquileia (Udine-Italia), In N. Poulou-Papadimitriou – E. Nodarou –
V. Kilikoglou (eds), LRCW 4. Late Roman Coarse Wares, Cooking
Wares and Amphorae in the Mediterranean. Archaeology and Archae-
ometry. The Mediterranean: A market without frontiers, BAR Int.
series 2616, Oxford, 943-953.
Chevalier, P. (1999) Les installations liturgiques des églises d’Istrie du
Ve au VIIe siècle, Histria Antiqua 5, 105-117.
Ciglenečki, S. (2009) Justinijanovo utvrďivanje Ilirika (Izvorni
znanstveni članak) / Justinian’s fortification of Illyricum, Archaeologia
Adriatica 3, 205-222.
Cipriano, M. T. (1994) La raccolta dei bolli sulle anfore italiche trovate
in Italia, In Epigrafia della produzione e della distribuzione. Actes de
la VIIe Rencontre franco-italienne sur l’épigraphie du monde romain
(Rome, 5-6 juin 1992), CEFR 193, 205-218.
Cipriano, S. (2003) Il consumo di derrate ad Altinum tra I secolo a.C. e
II secolo d.C.: i dati dei contenitori da trasporto, In Altinum. Studi
di archeologia, epigrafia e storia 3. Produzioni, merci e commerci in
Altino preromana e romana, Studi e Richerche sulla Gallia Cisalpina
17, Rome, 235-259.
Cipriano, S. (2008) Nuovi dati sulle anfore olearie istriane da Iulia Con-
cordia, In P. Basso – A. Buonopane – A. Cavarzere – S. Pesavento
Mattioli (eds), Est enim ille flos Italiae: Vita economica e sociale nella
Cisalpina romana, Verona, 303-312.
Cipriano, S. (2009) Le anfore olearie Dressel 6B, In S. Pesavento Mat-
tioli – M.-B. Carre (eds), Olio e pesce in epoca romana. Produzione
e commercio nelle regioni dell’Alto Adriatico. Atti del convegno
(Padova, 16 febbraio 2007), Antenor Quaderni 15, Roma, 173-189.
Cipriano, M. T. – Carre, M.-B. (1987) Note sulle anfore conservate
nel museo di Aquileia, Antichità Altoadriatiche 29, 479-494.
Cipriano, M. T. – Carre, M.-B. (1989) Production et typologie des
amphores sur la Côte Adriatique de l’Italie, In Amphores romaines et
histoire économique dix ans de recherche, CEFR 114, 67-104.
Cipriano, S. – Ferrarini, F. (2001) Le anfore romane di Opitergium,
Cornuda (Treviso), Venezia.
Cipriano, S. – Mazzocchin, S. (1998) I bolli di C. Laecanius Bassus:
Un aggiornamento alla luce di nuovi dati da Patavium, Aquileia Nostra
69, 361-378.
Cipriano, S. – Mazzocchin, S. (2000) Considerazioni su alcune
anfore Dressel 6B bollate. I casi di VARI PACCI e PACCI, APICI
e APIC, P.Q.SCAPVLAE, P.SEPVLLI.F e SEPVLLIVM, Aquileia
Nostra 71, 149-192.
Cipriano, S. – Mazzocchin, S. (2002) Analisi di alcune serie bollate
de anfore Dressel 6B (AP.PVLCHRI, FLAV.FONTAN e FONTANI,
L.IVN.PAETINI, L.TRE.OPTATI), Aquileia Nostra 73, 305-340.
Cipriano, S. – Mazzochin, S. (2011) Bonifiche con anfore a Padova:
note di aggiornamento alla cronologia e alla distribuzione topografica,
Antenor Quaderni 20, 331-367.
Cipriano, S. – Mazzochin, S. (2012) Produzioni anforarie dell’Ita-
18.03.19 14:40
 BIBLIOGRAPHY
lia alto e medioadriatica in età romana, In S. Fiorello (ed), Ceramica
romana nella Puglia Adriatica, Bari, 241-254.
Cipriano, S. – Mazzocchin, S. – De Vecchi, G. – Zanco, A. (2005)
Le anfore ad impasto grezzo rinvenute nella Venetia: tipologia, cro-
nologia, distribuzione, caratteri chimico-petrografici e tecnologia di
produzione, In P. Borgard – J. P. Brun – M. Picon (eds), L’Alun de
Méditerranée, Naples – Aix-en-Provence, 187-196.
Cipriano, S. – Mazzochin, S. – Pastore, P. (1997) Nuove considera-
zioni sui commerci del territorio patavino in età imperiale. Analisi di
alcune tipologie di anfore da recenti scavi, QuadAVen 13, 99-109.
Cirelli, E. (2007) Ravenna e il commercio nell’adriatico in età tardoan-
tica, In Felix Ravenna. La croce, la spada, la vela: l’alto Adriatico fra V e
VI secolo. Catalogo della Mostra, Milan, 45-50.
Cirelli, E. (2014) Typology and diffusion of Amphorae in Ravenna and
Classe between the 5th and the 8th centuries AD, In N. Poulou-Papa­
dimitriou – E. Nodarou – V. Kilikoglou (eds), LRCW 4. Late Roman
Coarse Wares, Cooking Wares and Amphorae in the Mediterranean.
Archaeology and Archaeometry. The Mediterranean: A market with-
out frontiers, BAR Int. series 2616, Oxford, 541-546.
Cirelli, E. – Cannavicci, A. (2014) A 6th century dump from Classe
(Ravenna), In N. Poulou-Papadimitriou – E. Nodarou – V. Kiliko-
glou (eds), LRCW 4. Late Roman Coarse Wares, Cooking Wares and
Amphorae in the Mediterranean. Archaeology and Archaeometry.
The Mediterranean: A market without frontiers, BAR Int. series 2616,
Oxford, 963-974.
Contino, A. – D’Alessandro, L. (2014) Materiali ceramici dagli scavi
della porticus Aemilia (Testaccio, Roma). Campagna di scavo 2011-
2012, ReiCretActa 43, 323-334.
Connelly, J. B. – Wilson, I. (2002) Hellenistic and Byzantine cisterns
on Geronisos Island, Report Dep. Ant. Cyprus, Nicosia, 269-292.
Cunja, R. (1996) Capodistria tardoromana e altomedievale. Lo scavo
archeologico nell’ex Orto dei Capuccini negli anni 1986-1987 alla luce
dei reperti dal V al IX secolo d.C., Koper.
D’Incà, C. – Kovačić, V. – Marchiori, A. – Marion, Y. – Rosada,
G. – Rousse, C. – Tassaux, F. – Zabeo, M. (2010) Loron-Lorun
(Parenzo-Porec, Istria). Una villa marittima nell’agro parentino: la
campagna di ricerca 2009, Histria Antiqua 19, 313-327.
Degrassi, A. (1953) Aquileia e l’Istria in età romana, Studi Aquileiesi
offerti a G. Brusin, Aquileia, 51-65, reprint In Scritti vari di antichità II
(Roma 1962) 951-963.
Degrassi, V. – Gaddi, D. – Mandruzzato, L. (2007) Amphorae and
coarse ware from Late Roman-Early medieval layers of the recent exca-
vations in Tergeste/Trieste (Italy), In M. Bonifay – J.-C. Tréglia (eds),
LRCW 2. Late Roman Coarse Wares, Cooking Wares and Amphorae
in the Mediterranean. Archaeology and Archaeometry, BAR Int. series
1662, Oxford, 503-510.
Degrassi, V. – Gaddi, D. – Mandruzzato, L. – Maselli Scotti,
F. (2010) Tergeste (Trieste-Italia), Materiale ceramico dai livelli di
costruzione delle mura tardo imperiali (prima metà V sec. d.C.), In
S. Menchelli – S. Santoro – M. Pasquinucci – G. Guiducci (eds),
LRCW 3. Late Roman Coarse Wares, Cooking Wares and Amphorae
in the Mediterranean. Archaeology and Archaeometry. Comparison
between western and eastern Mediterranean, BAR Int. series 2185,
Oxford, 579-584.
Degrassi, V. – Maggi, P. (2011) Il pozzo occidentale del foro di Aqui-
leia. Commerci e consumi attraverso l’evidenza dei materiali di riempi-
mento, In Archeologia e tecnica dei pozzi per acqua dalla pro-protostoria
all’età moderna, Trieste, 261-270.
Delattre, A. L. (1894) Le mur à amphores de la colline Saint-Louis à
Carthage, Bulletin archéologique du Comité des travaux historiques et
scientifiques, 89-119.
BEZECZKY_Text.indd 227
227
Demesticha, S. (2005) Late Roman Amphora 13: a study of diversifica-
tion, In Trade Relations in the Eastern Mediterranean from the Late
Hellenistic Period to Late Antiquity: The Ceramic Evidence, Odense,
169-178.
Desy, P. (1989) Les timbres amphoriques de l’Apulie républicaine, BAR
Int. series 554, Oxford.
Di Stefano Manzella, I. (1996) C. Iulius Africanus eques Romanus,
orator, dominus figlinarum?, Zeitschrift für Papyrologie und Epigraphik
111, 211-217.
Diamanti, C. (2010a) Local production and import of amphoras at Hala-
sarna of Kos Island (5th-7thc.), Athens.
Diamanti, C. (2010b) Stamped Late Roman/Proto-Byzantine Ampho-
ras from Halanarsa of Kos, ReiCretActa 41, 1-8.
Diamanti, C. (2012) Byzantine Emperors on Stamped Late Roman/
Early Byzantine Amphoras, ReiCretActa 42, 1-5.
Diamanti, C. (2016) The Late Roman Amphora Workshops of Paros
Island in the Aegean Sea, ReiCretActa 44, 691-697.
Didioumi, S. (2014) Local pottery production in the island of Cos,
Greece from the Early Byzantine period. A preliminary report, In
N. Poulou-Papadimitriou – E. Nodarou – V. Kilikoglou (eds), LRCW
4. Late Roman Coarse Wares, Cooking Wares and Amphorae in
the Mediterranean. Archaeology and Archaeometry. The Mediter-
ranean: A market without frontiers, BAR Int. series 2616, Oxford,
169-180.
Dixneuf, D. (2011) Amphores égyptiennes. Production, typologie, con-
tenu et diffusion (IIIe siècle avant J.-C. - IXe siècle après J.-C.), Études
Alexandrines 22, Alexandria.
Dobreva, D. (2012) I materiali archeologici, In J. Bonetto –
A. R. Ghiot­to (eds), Aquileia-Fondi ex Cossar. Missione archeologica
2012, Padova, 101-136.
Dobreva, D. – Riccato, A. (2015) Aquileia e il vicino oriente. Il com-
mercio di prodotti levantini in area adriatica, Aquileia Nostra 86, 111-
139.
Dore, J. (2001) The major pottery deposits following the disuse of the
East Baths, In L. Stirling – D. Mattingly – N. Ben Lazreg (eds), Lepti-
minus (Lamta). Report no 2, JRA Suppl. 41, 75-98.
Dugonjić, P. (2015) Rhodian amphorae in the Adriatic: a preliminary
report on the evidence from Croatia, In S. Demesticha (ed), Per Ter-
ram, per Mare. Seaborne Trade and the Distribution of Roman Ampho-
rae in the Mediterranean, Uppsala, 245-256.
Durn, G. (2003) Terra Rossa in the Mediterranean Region: Parent mate-
rials, composition and origin, Geologia Croatica 56/1, 83-100.
Durn, G. – Ottner, F. – Slovenec, D. (1999) Mineralogical and geo-
chemical indicators of the polygenetic nature of terra rossa in Istria,
Croatia, Geoderma 91, 125-150.
Duval, N. (1974) L’architecture chrétienne de l’Afrique du Nord dans
ses rapports avec le Nord de l’Adriatique, In Aquileia e l’Africa. Atti
della IV. Settimana di studi aquileiesi. 28 aprile- 4 maggio 1973, Udine,
353-368.
Džin, K. – Šalov, T. (2008) The latest archaeological investigations in
front of the amphitheatre in Pola, 2007-2008, Exhibition catalogue, Pola.
Džin, K. (2011) Rimske vile i uvjeti stanovanja na pulskom ageru. Neki
primjeri, Histria Antiqua 20, 31-107.
Egger, R. (1916) Frühchristliche Kirchenbauten im südlichen Norikum,
SoSchrÖAI 9, Wien, 122.
Egloff, M. (1977) Kellia: La poterie copte. Quatre siècles d’artisanat et
d’échanger en Basse-Égypte, Geneve.
Ehmig, U. (2003) Die römischen Amphoren aus Mainz, Frankfurter
Archäologische Schriften 4, Möhnesee.
Ehmig, U. (2007) Die römischen Amphoren im Umland von Mainz,
Frankfurter Archäologische Schriften 5, Wiesbaden.
18.03.19 14:40
 228 BIBLIOGRAPHY
Empereur, J.-Y.– Hesnard, A. (1987) Les amphores hellénistiques du
monde égéen, In P. Lévêque – J.-P. Morel (eds), Céramiques hellénis-
tiques et romaines 2, Paris, 9-71.
Fabião, C. (2008) Las ánforas de Lusitania, In D. Bernal – A. Ribera i
Lacomba (eds), Cerámicas hispanorromanas. Un estado de la cuestión,
Cadiz, 725-745.
Facchini, G. M. (1997) Anfore da Calvatone Romana. Osservazioni
sulle vie commerciali in area mediopadana, In Acme. Annali della
Facoltà di Lettere e Filosofia dell’Università degli Studi di Milano 50 II,
Milan, 39-58.
Fernández Fernández, A. (2016) Lusitanian Amphorae in the
Northwest of the Iberian Peninsula, In I. Vaz Pinto – R. R. De
Almeida – A. Martin (eds), Lusitanian Amphorae: Production and
Distribution, Roman and Late Antique Mediterranean Pottery 10,
Oxford, 273-284.
Ferrarini, F. (1993) Osservazioni su due tipologie di anfore della media
età imperiale da Altino, QuadAVen 11, 157-164.
Ferrazzoli, A. F. (2010) Economy of Roman Eastern Rough Cilicia:
Some Archaeological Indicators, In H. di Giuseppe – M. dalla Riva
(eds), 17th International Congress of Classical Archaeology. Meetings
between Cultures in the Ancient Mediterranean, Bollettino di archeo-
logia on line, volume speciale, Rome, 39-50.
Filis, K. (2016), The local transport amphorae from Aigion, In S. Japp –
P. Kögler (eds), Traditions and Innovations: Tracking the Development
of Pottery from the Late Classical to the Early Imperial Periods, Berlin,
November 2013, 7th-10th , Wien, 151-167.
Formenti, F. – Hesnard, A. – Tchernia, A. (1978) Une amphore
“Lamboglia 2” contenant du vin dans l’épave de la Madrague de Giens,
Archaeonautica 2, 95-100.
Franco, P. (2012) African amphorae from Portus, Unpublished PhD dis-
sertation, University of Southampton.
Furlan, A. (1993) Censimento dei bolli laterizi di un’area campione
a nord-est di Aquileia, In C. Zaccaria (ed), I laterizi di età romana
nell’area nordadriatica, Rome, 199-205.
G.A.V.O. (1993) Gruppo Archeologico del Veneto Orientale, Nuovi
laterizi bollati dalla parte meridionale dell’agro di Iulia Concordia,
In C. Zaccaria (ed), I laterizi di età romana nell’area nordadriatica,
Rome, 207-211.
Gabucci, A. – Quiri, E. (2008) Eporedia: Appunti su terre sigillate e
anfore tra tarda repubblica ed età imperiale, Quaderni della Soprinten-
denza Archeologica del Piemonte 23, 45-78.
Gaddi, D. – Degrassi, V. (2016) Lusitanian Amphorae in the North-
ern Adriatic Region: the Western Part of the Decima Regio, In I. Vaz
Pinto – R. R. De Almeida – A. Martin (eds), Lusitanian Amphorae:
Production and Distribution, Roman and Late Antique Mediterra-
nean Pottery 10, Oxford, 437-444.
Gandolfi, D. – Murialdo, G. – Capelli, C. – Bonifay, M. (2010)
Anfore africane di tardo V-VII secolo in Liguria (Italia): un aggiorna-
mento dei dati archeologici e archeometrici, In S. Menchelli – S. San-
toro – M. Pasquinucci – G. Guiducci (eds), LRCW 3. Late Roman
Coarse Wares, Cooking Wares and Amphorae in the Mediterranean.
Archaeology and Archaeometry. Comparison between western and
eastern Mediterranean, BAR Int. series 2185, Oxford, 33-56.
García Brosa, G. – Ozcáriz Gil, P. (2007) Los grafitos nominales de
las ánforas Dressel 20: El caso del grafito ‘Vitalis’, In Acta XII Congres-
sus Internationalis Epigraphiae Graecae et Latinae, Barcelona, 549-554.
García Vargas, E. (1998) La producción de ánforas en la bahía de Cádiz
en época romana (siglos II aC-IV dC), Écija.
García Vargas, E. (2000) La producción de ánforas “romanas” en el
sur de Hispania. República y Alto Imperio, In D. Bernal – E. García
Vargas (eds), Ex Baetica Amphorae. Conservas, aceite y vino de la Bética
en el Imperio Romano I, Écija, 57-174.
BEZECZKY_Text.indd 228
García Vargas, E. (2016) Amphora Circulation in the Lower Gua-
dalquivir Valley in the Mid Imperial Period: the Lusitana 3 Type,
In I. Vaz Pinto – R. R. De Almeida – A. Martin (eds), Lusitanian
Amphorae: Production and Distribution, Roman and Late Antique
Mediterranean Pottery 10, Oxford, 285-298.
García Vargas, E. – De Almeida, R. – González Cesteros,
H. (2011) Los tipos anfóricos del Guadalquivir en el marco de los
envases hispanos del siglo I a.C. Un universo heterogéneo entre la imi-
tación y la estandarización, Spal 20, 185-283.
Gelichi, S. – Negrelli, C. (2008) Anfore e commerci nell’alto Adria-
tico tra VIII e IX secolo, MEFRM 120-2, Rome, 307-326.
Gerousi, E. (2014) A Late Roman Workshop at Dilesi in Boeotia,
In N. Poulou-Papadimitriou – E. Nodarou – V. Kilikoglou (eds),
LRCW 4. Late Roman Coarse Wares, Cooking Wares and Ampho-
rae in the Mediterranean. Archaeology and Archaeometry. The
Mediterranean: A market without frontiers, BAR Int. series 2616,
Oxford, 193-202.
Girardi Jurkić, V. (1985) Brioni, Castrum Bizantino, In V. Girardi
Jurkić (ed), Archeologia e Arte dell’Istria, Pula, 91-94.
Glaser, F. (1997) Reliquiengräber – Sonderbestattungen der Spätantike,
Arheološki Vestnik 48, 231-246.
Glicksman, K. (2005) Internal and external trade in the Roman prov-
ince of Dalmatia, Opvscvla Archaeologica 29, 189-230.
Gnirs, A. (1904) Antike Funde aus Pola und Umgebung, ÖJh 7, Beibl.
131-146.
Gnirs, A. (1906) Forschungen im südlichen Istrien, ÖJh 9, Beibl. 25-48.
Gnirs, A. (1908a) Istrische Beispiele für Formen der antik-römischen
villa rustica, Jahrbuch für Altertumskunde 2, 124-143.
Gnirs, A. (1908b) Forschungen im südlichen Istrien, ÖJh 11, Beibl. 167-
186.
Gnirs, A. (1910a) Eine römische Tonwarenfabrik in Fasana bei Pola,
Jahrbuch für Altertumskunde 4, 79-88.
Gnirs, A. (1910b) Forschungsergebnisse aus dem südlichen Istrien, ÖJh
13, Beibl. 95-106.
Gnirs, A. (1911a) Baudenkmale aus der Zeit der oströmischen Herr-
schaft auf der Insel Brioni grande, Jahrbuch für Altertumskunde 5,
75-97.
Gnirs, A. (1911b) Fasana: Neue Funde aus der Figlina des C. Laekanius
Bassus, ÖJh 14, 35-39.
Gnirs, A. (1911c) Frühe christliche Kultanlagen im südlichen Istrien,
Jahrbuch des Kunsthistorischen Institutes der k. u. k. Zentral-Kommis-
sion für Denkmalpflege 5, Wien, 38-48.
Gnirs, A. (1911d) Grabungen im Gebiet der antiken Herrschaftsvilla
von Val Bandon, ÖJh 14, 155-186.
Gnirs, A. (1915) Forschungen über antiken Villenbau in Südistrien, ÖJh
18, Beibl. 99-164.
Gomezel, C. (1996) I laterizi bollati romani del Friuli-Venezia Giulia:
analisi, problemi e prospettive, Gruppo Archeologico Veneto Orientale,
Portogruaro.
González, M. – Lagostena, L. – Modrzewska, I. – Oddone,
M. – Pianetti, F. (2000) L’importazione del garum iberico a Verona,
In Ex Baetica amphorae, Écija, 837-852.
González Cesteros, H. (2010) La llegada de ánforas hispanas a
Germania durante los últimos siglos de la dominación romana. Una
cuestión para el future, CuPaUAM 36, 107-129.
González Cesteros, H. (2012) Scomber gaditano en Éfeso. Una
Dressel 12 con titulus encontrada en la “Casa Aterrazada 2” de Éfeso,
Dialogues d’histoire ancienne 38/I, 9-22.
González Cesteros, H. (2015) Hallazgos de productos tarraconen-
ses en la frontera germana. Un mercado secundario, In COMAMF. La
difusió comercial de les àmfores vinàries produïdes al nord-est peninsular
(s. I aC-I dC), Barcelona, 205-220.
18.03.19 14:40
 BIBLIOGRAPHY
González Cesteros, H. – Yilmaz, Z. (2013) Dos nuevas piezas
hispanas encontradas en Éfeso, Boletín de la SECAH 4, 23-25.
González Cesteros, H. – Yilmaz, Z. (forthcoming) Amphoren-
funde aus der Hafennekropole, In M. Steskal (ed), Forschungen in der
Hafennekropole von Ephesos. Die Grabungen seit 2005 im Bereich
des Hafenkanals, SoSchrÖAI, Vienna.
González Cesteros, H. – Almeida, R. – Curbera, J. (2016) Special
Fish Products for the Jewish Community? A Painted Inscription on a
Beltrán 72 Amphorae from Augusta Emerita (Mérida, Spain), Herom.
Journal on Hellenistic and Roman Material Culture 5, 197-236.
Gostenčnik, K. (2002) Amphoren mit Laecanius-Stempel aus Ober­
drauburg und Baldersdorf, Carinthia I 192, 165-171.
Grace, V. R. (1949) Standard pottery containers of the ancient Greek
world, In Commemorative Studies in Honor of Theodore Leslie
Shear, Hesperia Suppl. 8, Princeton, 175-189.
Grace, V. R. (1965) The commercial amphoras from the Antikythera
Shipwreck, Transactions of the American Philosophical Society 55/3,
5-17.
Grace, V. R. (1979) Amphoras and ancient wine trade, Excavations of
the Athenian Agora, Picture Book No. 6, Athens.
Grace, V. R. – Savatianou-Pétropoulakou, M. (1970) Les timbres
amphoriques grecques, In Délos 27, Paris, 277-382.
Gregorutti, C. (1888) Le marche di fabbrica dei laterizi di Aquileia,
Archeografo Triestino 14, Trieste, 345-367.
Gugl, C. (2003) Keramik, In A. Galik – C. Gugl – G. Sperl, Feldkirchen
in Kärnten. Ein Zentrum norischer Eisenverhüttung, DenkschrWien
314, Wien, 43-45.
Gugl, C. (2004) Amporen mit Stempel und Tituli picti aus dem Amphi-
teather, In R. Jernej – C. Gugl (eds), Virunum, Das römische Amphi-
theater, Die Grabungen 1998-2001, Klagenfurt, 40-42.
Gulam, V. – Pollak, D. – Podolszki, L. (2014) The analysis of the
flysh badlands inventory in central Istria, Croatia, Geologia Croatia
67/1, 1-15.
Hayes, J. W. (1991) Paphos III. The Hellenistic and Roman Pottery, Nico-
sia.
Hayes, J. W. (1992) Excavations at Saraçhane in Istanbul II. The Pottery,
Princeton.
Heger, N. (1986) Frührömische Amphoren aus der Stadt Salzburg
(Mozartplatz 4), BayVgBl 51, 131-161.
Hill, S. (1996) The Early Byzantine Churches of Cilicia and Isauria, Bir-
mingham Byzantine and Ottoman monographs 1, Birmingham.
Holwerda, J. H. – Braat, W. C. (1946) De Holdeurn bij Berg en Dal,
centrum van pannenbakkerij en aardewerkindustrie in den Romein-
schen tijd, Internationales Archiv für Ethnographie, 45 Suppl., Leiden.
Horvat, J. – Bavdek, A. (2009) OKRA, Vrata med Sredozemljem in
Srednjo Evropo / OCRA The gateway between the Mediterranean and
Central Europe, Ljubljana.
Hoti, A. – Capelli, C. – Piazza, M. (2008) Gli spatheia di Dyrra-
chium, ReiCretActa 40, 1-8.
Hoxha, G. (2003) Scodra dhe Praevalis në antikitetin e vonë, Shkodra.
Huguet, C. (2012) La céramique et les matériaux de construction en
terre cuite, In G. Boetto – I. Radić Rossi – S. Marlier – Z. Brusić (eds),
L’épave de Pakoštane, Croatie (fin IVe - début Ve s. apr. J.-C.). Résultats
d’un projet de recherche Franco-Croate, Archaeonautica 17, 128-132.
Isler, H. P. (1969) Heraion von Samos: Eine frühbyzantinische Zisterne,
AM 84, 202-236.
Jaritz, H. (1982) Abū Mīna. Zehnter vorläufiger Bericht. Kampagnen
1980 und 1981, MDAIK 38, 145-148.
Johnson, B. L. (2008) The Pottery, In J. Patrich (ed), Archaeologi-
cal Excavations at Caesarea maritime. Areas CC, KK and NN. Final
Reports 1, Jerusalem, 13-206.
BEZECZKY_Text.indd 229
229
Józsa, S. – Szakmány, Gy. (1987) Petrology, In T. Bezeczky (ed),
Roman amphorae from the Amber Route in Western Pannonia, BAR
Int. series 386, Oxford, 103-127.
Józsa, S. – Szakmány, Gy. – Obbágy, G. – Bendő, Zs. – Taubald,
H. (2016) A fažanai (Isztria, Horvátország) Laecanius amphorák
archeometriája / Archaeometry of Laecanian amphorae from Fažana,
Istria, Archeometriai Műhely, www.ace.hu/am 2016. XIII. 2, 95-130.
Jurišić, M. (2000) Ancient Shipwrecks of the Adriatic. Maritime trans-
port during the first and second centuries AD, BAR Int. series 828,
Oxford.
Kapitän, G. (1972) Le anfore del relitto romano di Capo Ognina (Sir-
acusa), Recherches sur les amphores romaines, CEFR 10, Rome, 243-
252.
Kara, Ü. (2015) Yenikapı Limanı’nda Ele Geçen Damgalı Amphoralar,
In A. Diler – K. Şenol – Ü. Aydinonglu (eds), Olive oil and wine pro-
duction in eastern Mediterranean during Antiquity, Izmir, 243-250.
Karagiourgou, O. (2001) LRA 2: a container for the Military annona
on the Danubian border? In M. Decker – S. A. Kingsley (eds), Econ-
omy and exchange in the East Mediterranean during late antiquity,
Oxford, 129-166.
Kassab Tezgör, D. (2010) Les fouilles et le materiel de l’atelier
amphorique de Demirci près de Sinope, Varia Anatolica 22, Istanbul.
Keay, S. J. (1984) Late Roman Amphorae in the Western Mediterranean.
A Typology and Economic Study: the Catalan Evidence, BAR Int.
series 196, Oxford.
Kirigin, B. – Katunarić, T. – Šešelj, L. (2006) Preliminary notes on
some economic and social aspects of amphorae and fine ware pottery
from central Dalmatia, 4th-1st B.C., In Rimini e l’ Adriatico nell’ età
delle guerre puniche. Atti del Convegno Internazionale di Studi Rimini,
Musei Comunali, 25-27 marzo 2004, Bologna, 191-225.
Križman, M. (1979) Antička svjedočanstva o Istri, Pula – Rijeka.
Ladstätter, S. (2005) Die Keramik, In H. Thür (ed), Hanghaus 2 in
Ephesos. Die Wohneinheit 4. Baubefund, Ausstattung, Funde, FiE
VIII/6, Wien, 230-358.
Ladstätter, S. (2008) Römische, spätantike und byzantinische Kera-
mik, In M. Steskal – M. La Torre (eds), Das Vediusgymnasium in
Ephesos. Archäologie und Baubefund, FiE XIV/1, Wien, 97-189.
Lafon, X. (2001) Villa maritima. Recherches sur les villas littorales de
l’Italie Romaine. 3. siècle av. J.-C. / 3. siècle ap. J.-C., Bibliothèque de
écoles françaises d’Athènes et de Rome 307, 459.
Lamboglia, N. (1955) Sulla cronologia delle anfore romane di età
repubblicana (II-I secolo a.C.), Rivista di Studi Liguri 21, 241-270.
Laubenheimer, F. (2015) Les circuits d’exportation des vins de Tar-
raconaise en Gaule, In COMAMF. La difusió comercial de les àmfores
vinàries produïdes al nord-est peninsular (s. I aC-I dC), Barcelona, 181-
192.
Lindhagen, A. (2009) The transport amphoras Lamboglia 2 and Dres-
sel 6A. A central Dalmatian origin?, JRA 22, 1, 83-108.
Lindhagen, A. (2013) The freedmen milieus at Delos and Novara. New
perspectives on the Lamboglia 2 wine trade, In A.-L. Schallin (ed),
Perspectives on Ancient Greece, Papers in celebration of the 60th anni-
versary of the Swedish Institute at Athens, Acta Instituti Atheniensis
Regni Sueciae 8, 22, Stockholm, 231-250.
Liou, B. (1987) Inscriptions peintes sur amphores: Fos (suite), Marseille,
Toulon, Port-la-Nautique, Arles, Saint-Blaise, Saint-Martin-de-Crau,
Mâcon, Calvi, Archaeonautica 7, 55-139.
Lipovac Vrkljan, G. (2009) L’officina ceramica di Crikvenica, In
S. Pesavento Mattioli – M.-B. Carre (eds), Olio e pesce in epoca
romana. Produzione e commercio nelle regioni dell’Alto Adriatico.
Atti del convegno (Padova, 16 febbraio 2007), Antenor Quaderni 15,
Roma, 309-314.
18.03.19 14:40
 230 BIBLIOGRAPHY
Lipovac Vrkljan, G. (2011) Lokalna keramičarska radionica Seksta
Metilija Maksima u Crikvenici – Crikveničke amfore ravnog dna / Local
pottery workshop of Sextus Metilius Maximus in Crikvenica – Crok­
venica flat bottomed amphorae, In G. Lipovac Vrkljan – I. Radić Rossi –
B. Šiljeg (eds), Rimske keramičarke I staklarske radonice. Proizvodnja I
trgovina na Jadranskom prostoru. Zbornik I. Medunarodnog Archeološkog
Kolokvija, Crikvenica, 23.-24. Listopada 2008, Crik­venica, 3-18.
Lund, J. (2000a) Transport Amphorae as Evidence of Exportation of
Italian Wine and Oil to the Eastern Mediterranean in the Hellenistic
Period, In J. Lund – P. Pentz (eds), Between Orient and Occident. Stud-
ies in Honour of P. J. Riis, Kopenhagen, 77-99.
Lund, J. (2000b) The “Pinched-handle” Transport Amphorae as Evidence
of the Wine Trade of Roman Cyprus, In Ιωαννίδη, Γ.Κ. – Χατζηστύλη,
Σ.Α. (eds), Πρακτικά τoυ τρίτoυ διεθvoύς κυπρoλoγικoύ συvεδρίoυ
(Λευκωσία, 16-20 Aπριλίoυ 1996), Τόμoς Α’, Λευκωσία, 565-578.
Lusuardi Sienna, L. (1990) Contenitori da trasporto, In Milano capi-
tale dell’impero romano 286-402 d. C., Milan, 379-385.
Maggi, P. – Marion, Y. (2007) Sénateurs et activités économiques:
l’enseignement des timbres de Loron (Croatie), In XII Congressus
Internationalis Epigraphiae Graecae et Latinae: Provinciae Imperii
Romani inscriptionibus descriptae Barcelona, 3-8 Septembris 2002, Bar-
celona, 857-862.
Maggi, P. – Urban, M. (2001) La zona settentrionale del porto fluviale
di Aquileia, In C. Zaccaria (ed), Strutture portuali e rotte marittime
nell’Adriatico di età romana, CEFR 280, Trieste – Rome, 245-259.
Maier-Maidl, V. (1992) Stempel und Inschriften auf Amphoren vom
Magdalensberg, Archäologische Forschungen zu den Grabungen auf
dem Magdalensberg 11, Klagenfurt.
Manacorda, D. (1995) A proposito delle anfore della Pannonia
romana: appunti e riflessioni, In La Pannonia e l’Impero Romano,
Annuario dell’ Accademia d’Ungheria, Roma, 177-191.
Manacorda, D. (2010) Il “Misterioso” MESCAE, Donne imprendit-
rici nell’Istria romana, ReiCretActa 41, 217-227.
Manacorda, D. (2012) Le produzioni, In D. Manacorda – S. Pallecchi
(eds), Le fornaci romane di Giancola (Brindisi), Bari, 103-292.
Mange, M. A. – Bezeczky, T. (2006) Petrography and Provenance of
Laecanius Amphorae from Istria, Northern Adriatic Region, Croatia,
Geoarchaeology 21-5, 429-460.
Mange, M. A. – Bezeczky, T. (2007) The provenance of paste and tem-
per in Roman amphorae from the Istrian Peninsula, Croatia, In M. A.
Mange – D. T. Wright (eds), Heavy Minerals in Use, Developments in
Sedimentology 58, 1007-1033.
Marangou-Lerat, A. (1995) Le vin et les amphores de Crète. De
l’époque classique à l’époque imperial, Études crétoises 30, Paris.
Mardešić, J. (1994) Amfore, In E. Marin (ed), Salona Christiana, Split,
293-296.
Marengo, S. M. (1981) Bolli laterizi di Quinto Clodio Ambrosio nel
Piceno, Picus 1, 105-113.
Marengo, S. – Paci, G. (2008) Per la circolazione delle anfore rodie e
tardo-repubblicane in area adriatica, In P. Basso – A. Buonopane –
A. Cavarzere – S. Pesavento Mattioli (eds), Est enim ille flos Italiae.
Vita economica e sociale nella Cisalpina romana, Verona, 313-328.
Marion, Y. – Starac, A. (2001) Les amphores, In F. Tassaux –
R. Matijašić – V. Kovačić, Loron (Croatie). Un grand centre de pro-
duction d’amphores à huile istriennes (Ier-IVe s. p.C.), Mémoires /
Ausonius 6, Bordeaux, 97-125.
Marion, Y. (2009) Les Dressel 6B de petites dimensions de Loron,
In S. Pesavento Mattioli – M.-B. Carre (eds), Olio e pesce in epoca
romana. Produzione e commercio nelle regioni dell’Alto Adriatico.
Atti del convegno (Padova, 16 febbraio 2007), Antenor Quaderni 15,
Roma, 281-287.
BEZECZKY_Text.indd 230
Martin, A. (2016) Lusitanian Amphorae at Ostia and in the Vesuvian
Region, In I. Vaz Pinto – R. R. De Almeida – A. Martin (eds), Lusi-
tanian Amphorae: Production and Distribution, Roman and Late
Antique Mediterranean Pottery 10, Oxford, 389-397.
Martin-Kilcher, S. (1987) Die römischen Amphoren aus Augst
und Kaiseraugst. Ein Beitrag zur römischen Handels- und Kulturge-
schichte 1. Die südspanischen Ölamphoren (Gruppe 1), Forschungen
in Augst 7/1, Augst.
Martin-Kilcher, S. (1993) Amphoren der späten Republik und der
frühen Kaiserzeit in Karthago. Zu den Lebensmittelimporten der
Colonia Iulia Concordia, RM 100, 269-320.
Martin-Kilcher, S. (2000) Les Laecanii et les amphores à huile d’Istrie,
JRA 13, 506-509.
Martin-Kilcher, S. (2005) Carthage: Imported eastern amphorae
in the Roman Colonia Iulia, In M. Berg Briese – L. Erik Vaag (eds),
Trade Relations in the Eastern Mediterranean from the Late Hellenis-
tic Period to Late Antiquity: The Ceramic Evidence, Halicarnassian
Studies 3, Odense, 202-220.
Marty, F. – Zaaraoui, Y. (2009) Contextes céramiques du Haut-
Empire de la bonification de l’Estagnon, à Fos-sur-Mer (Bouches-du-
Rhone), In L. Rivet – S. Saulnier (eds), Sites de production et de consom-
mation dans le Rhin supérieur, Marseille, 397-426.
Marušić, B. (1967) Kasnoantička i bizantska Pula / Das spätantike und
byzantinische Pula, Kulturhistorische Denkmäler in Istrien 6, Pula,
33-35.
Marušić, B. (1990) Še o istrski Kisi (Cissa) in kesenskem škofu
(episcopus Cessensis) / Noch einiges über das istrische Cissa und das
Kessensche Bistum (episcopatus Cessensis), Arheološki Vestnik 41,
403-424.
Maselli Scotti, F. – Degrassi, V. – Mian, G. (2003) Gli scarichi della
domus di Piazza Barbacan a Trieste, AttiMemIstria 103, 1, 19-105.
Matijašić, R. (1982) Roman Rural Architecture in the Territory of
Colonia Iulia Pola, AJA 86, 53-64.
Matijašić, R. (1983) Cronografia dei bolli laterizi della figulina pan-
siana nelle regioni adriatiche, MEFRA 95-2, 961-995.
Matijašić, R. (1985) Radionički žigovi na antičkim opekama zbirke
Arheološkog muzeja Istre, Jadranski zbornik 12, 287-303.
Matijašić, R. (1987) Vecchi e nuovi rivenimenti di tegole con bollo di
fabbrica in Istria, I - Istria Meridionale, Arheološki Vestnik 38, 161-192.
Matijašić, R. (1988) Ageri antičkih kolonija Pola i Parentium, Latina
et Graeca 6, Zagreb.
Matijašić, R. (1993a) Oil and Wine Production in Istria and Dalmatia
in Classical Antiquity and the Early Middle Ages, In M.-C. Amou­
retti – J.-P. Brun (eds), La production du vin et de huile en Méditer-
ranée / Oil and Wine Production in the Mediterranean Area. Actes
du Symposium International organisé par le Centre Camille Jullian et
le Centre Archéologique du Var, Aix-en-Provence et Toulon, 20-22
novembre 1991, BCH Suppl. 26, 255-258.
Matijašić, R. (1993b) Lo studio dei bolli laterizi romani in Istria dal
’700 ad oggi, In C. Zaccaria (ed), I laterizi di età romana nell’area
nord­adriatica, Rome, 127-133.
Matijašić, R. (1998) Gospodarstvo antičke Istre, Pula.
Matijašić, R. (2001) I porti dell’Istria e della Liburnia, In C. Zaccaria
(ed), Strutture portuali e rotte marittime nell’Adriatico di età romana,
CEFR 280, Trieste – Rome, 161-174.
Matijašić, R. (2008) Ostatci tijesaka u dvorištu rimske vile u uvali
Madona na Brijunima (tzv. Kastrum) / The remains of a press in the
court­yard of the Roman villa in Madona Bay on Brijuni Island (Kast-
rum), Archaeologia Adriatica 2, 289-300.
Matijašić, R. (2012) Povijest hrvatskih zemalja u kasnoj antici od Diokle-
cijana do Justinijana, Zagreb.
18.03.19 14:40
 BIBLIOGRAPHY
Maurina, B. (1999) Materiali aquileiesi di età romana al museo civico di
Rovereto (TN), Aquileia Nostra 70, 149-176.
Mayet, F. – Tavares Da Silva, C. (1998) L’atelier d’amphores de Pin-
heiro (Portugal), Paris.
Maza, G. – Silvino, T. – Dubreucq, E. (2014) Premières observations
sur les faciès céramiques de l’agglomération antique de Digoin (Saône-
et-Loire). Consommation et production/commercialisation?, In SFE-
CAG, Actes du Congrès de Chartres, Marseille, 799-828.
Mazzocchin, S. (2006) Gli apprestamenti con anfore, In S. Mazzoc-
chin – S. Tuzzato – S. Bonato – C. Rossi, Un nuovo drenaggio di
anfore dai Giardini dell’arena a Padova, Bollettino del Museo Civico di
Padova 95, 7-43.
Mazzocchin, S. (2007) Le anfore, In S. Mazzocchin – S. Tuzzato (eds),
Padova, via Acquette 9, nuovi dati dal settore meridionale della città
romana, QuadAVen 23, 123-139.
Mazzocchin, S. (2009) Le anfore con collo ad imbuto: Nuovi dati e
prospettive di ricerca, In S. Pesavento Mattioli – M.-B. Carre (eds),
Olio e pesce in epoca romana. Produzione e commercio nelle regioni
dell’Alto Adriatico. Atti del convegno (Padova, 16 febbraio 2007),
Antenor Quaderni 15, Roma, 191-213.
Mazzocchin, S. (2010) Le anfore riutilizzate nei drenaggi e i bolli, In
S. Mazzocchin – S. Tuzzato (eds), Padova, via dei Salici. Nota prelimi-
nare su un drenaggio a sud della città romana, QuadAVen 26, 21-29.
Mazzocchin, S. (2013) Vicenza, Traffici commerciali in epoca romana,
I dati delle anfore, Trieste.
Mazzotti, M. (1960) Gli altari paleocristiani degli edifici di culto raven-
nati, Corso di cultura sull’arte ravennate e bizantina, Ravenna, 238-246.
Mercando, L. (1974) Portorecanati (Macerata). La necropoli romana di
Portorecanati, Notizie degli Scavi dell’Antichità, 142-430.
Mercando, L. (1979) Portorecanati (Macerata), Notizie degli Scavi
dell’ Antichità, 180-280.
Migotti, B. (1992) Zusatz zur Datierung der außerstädtischen früh-
christlichen Architektur des breiteren salonitischen Bereiches,
Arheološki Vestnik 43, 111-133.
Mikes, T. – Frisch, W. – Eynatten, H. (2006) Geochemistry of
Eocene flysch sandstones in the MW External Dinarides, Acta Geo­lo­
gica Hungarica 49/2, 103-124.
Milotić, I. – Petrak, M. (2012) Roman Inscription from Materija
(Istria). Legal Relations and Disputes Between Local Community of
Rundictes and Senator Gaius Laecanius Bassus, Lex localis, Journal of
local self-government 10-4, 297-310.
Mirabella Roberti, M. (1935) Notiziario archeologico (1935-1936),
AttiMemIstria 47, 285-307.
Miškec, A. (2002) Die Fundmünzen der römischen Zeit in Kroatien,
Abteilung 17, Istrien, Mainz.
Mlakar, Š. (1975-76) Fortifikacijska arhitektura na otoku Brioni.
Bizantski kastrum / Fortifikationsarchitektur auf Brioni. Byzantini­
sches Kastrum, Histria Archaeologica 6-7, 10-34.
Mlakar, Š. (1979) Neki prilozi poznavaj arheološke topografije Istre /
Eine Beilage zur archäologischen Topographiekunde Istriens, Histria
Archaeologica 10-2, 19-45.
Mócsy, A. – Feldmann, R. – Marton, E. – Szilagyi, M. (1983)
Nomenclator, Dissertationes Pannonicae III, 1, Budapest.
Mohorovičić, A. (1957) Problem tipološke klasifikacije objekata Sred-
njov-jekovne arhitekture na području Istre i Kvarnera, Ljetopis JAZU
62, 491-493.
Mongardi, M. (2016) Anfore di morfologia betica con iscrizioni
dipinte dalla regio VIII Aemilia, In D. Járrega Domínguez – P. Berni
Millet – H. González Cesteros (eds), III Congreso Internacional de la
SECAH - Ex Officina Hispana, Madrid, 557-568.
Monsieur, P. (2001) Note préliminaire sur les amphores découvertes à
Pessinonte (Annexe I), Anatolia Antiqua 9, 73-84.
BEZECZKY_Text.indd 231
231
Monsieur, P. (2007) Graffiti, stempels en tituli picti op amforen: waar-
devolle informatiedragers voor de kennis van antieke economie en
naamkunde, Handelingen der Koninklijke Zuid-Nederlandse Maats-
chappij voor Taal- en Letterkunde en Geschiedenis 60, 101-129.
Monsieur, P. – Verreyke, H. – Vermeulen, F. (2006) Processing
pottery of the Potenza Valley Survey (Marche, Italy), In D. Malfitana –
J. Poblome – J. Lund (eds), Old pottery in a new century. Innovating
perspectives on Roman pottery studies. Atti del Convegno Internazionale
di Studi Catania, 22-24 Aprile 2004, Catania, 309-314.
Mora Serrano, B. – Corrales Aguilar, P. (1997) Establecimien-
tos salsarios y producciones anfóricas en los territorios malacitanos, In
Figlinae Malacitanae. La producción de cerámica romana en los territo-
rios malacitanos, Malaga, 27-59.
Morais, R. – Fabião, C. (2007) Novas produções de fabrico lusitano:
problemáticas e importância económica, In Cetariae 2005. Salsas y
Salazones de Pescado en Occidente durante la Antigüedad, BAR Int.
series 1686, Oxford, 127-133.
Moros Díaz, J. – Berni Millet, P. (2011) Novedades sobre epigrafía
anfórica bética para la zona productora de Adelfa (Carmona, Sevilla),
Boletín ‘Ex Officina Hispana’ 3, 34-45.
Mukai, T. (2014) Production vs. consumption in the Northern gulf of
Hammamet (Tunisia): Tefernine and Sidi Jdidi, ReiCretActa 43, 607-
616.
Mukai, T. (2016) La céramique du groupe épiscopal d’Aradi/Sidi
Jdidi (Tunisie), Roman and Late Antique Mediterranean Pottery 9,
Oxford.
Nacef, J. (2007) Nouvelles données sur l’atelier de potiers de Henchir
ech Chekaf (Ksour Essef, Tunisie), In M. Bonifay – J.-C. Tréglia (eds),
LRCW 2. Late Roman Coarse Wares, Cooking Wares and Amphorae
in the Mediterranean. Archaeology and Archaeometry, BAR Int. series
1662, Oxford, 581-591.
Nacef, J. (2014) Nouveaux témoignages sur la production de la céramique
antique du Sahel tunisien, In N. Poulou-Papadimitriou – E. Noda-
rou – V. Kilikoglou (eds), LRCW 4. Late Roman Coarse Wares,
Cooking Wares and Amphorae in the Mediterranean. Archaeology
and Archaeometry. The Mediterranean: A market without frontiers,
BAR Int. series 2616, Oxford, 235-253.
Nacef, J. (2015a) La production de la céramique antique dans la région
de Salakta et Ksour Essef (Tunisie), Roman and Late Antique Mediter-
ranean Pottery 8, Oxford, 246-251.
Nacef, J. (2015b) Un atelier de potier à la périphérie de la ville de Thapsus
aux premiers siècles de l’Empire, Antiquités Africaines 51, 40-43.
Negrelli, C. (2008) Rimini Capitale. Strutture insediative, sociali ed eco-
nomiche tra V e VIII secolo, Florence.
Neuru, L. (1990) Appendix 2: The Pottery of the Kasserine Survey, In
B. Hitchner, The Kasserine Archeological Survey, 1987, Antiquités
Africaines 26, 231-260.
Oikonomidès, N. (1989) Commerce et production de la soie à Byz-
ance, In Hommes et richesses dans l’Empire byzantin (IVe-VIIe siècle).
Tome I, Paris, 187-192.
Opait, A. (1997-98) North African and Spanish Amphorae in Scythia
Minor, Il Mar Nero. Annali di archeologia e storia 3, 47-95.
Opait, A. − Diamanti, C. (2014) Imperial Stamps on Early Byzantine
Amphoras. The Athenian Agora Exemples, ReiCretActa 43, 55-61.
Opait, A. − Tsaravopoulos, A. (2011) Amphorae of Dressel 24
Similis Type in the Central Aegean Area (Chios-Erythrai-Kyme),
Papers of the British School at Athens 106, 276-323.
Oren-Pascal, M. – Bernal Casasola, D. (2000) Ánforas sudhispanas
en Caesarea Maritima. Un ejemplo de importación de vino, aceite y
conservas de pescado béticas en Iudea, In D. Bernal – E. García Vargas
(eds), Ex Baetica Amphorae. Conservas, aceite y vino de la Bética en el
Imperio Romano III, Écija, 989-1012.
18.03.19 14:40
 232 BIBLIOGRAPHY
Oxé, A. – Comfort, H. (1968) Corpus vasorum Arretinorum: a cata-
logue of the signatures, shapes and chronology of Italian sigillata, Bonn.
Oxé, A. – Comfort, H. (2000) Corpus vasorum Arretinorum: a cata-
logue of the signatures, shapes and chronology of Italian sigillata, second
edition by Ph. Kenrick, Bonn.
Paić, A. – Bulić, D. (2008) Katalog, In Fažana ispod pločnika: izložba /
Sotto il lastricato di Fasana: mostra, Pula, 17-40.
Pais, H. (1884) Corpus Inscriptionum Latinarum Supplementa Italica,
Galliae Cisalpinae Fasc. 1, Roma.
Pacetti, F. (1998) La questione delle Keay 52 nell’ambito della produzi-
one anforica in Italia, In L. Saguì (ed), Ceramica in Italia: VI-VII secolo,
Florence, 185-208.
Palágyi, S. (2007) Nachlass der Wein- und Ölproduktion/Handel der
Römerzeit in Pannonien, besonders in Baláca, Histria Antiqua 15,
281-292.
Panella, C. (1970) Le Anfore, Studi Miscellanei 16, Ostia II, Roma, 102-
156.
Panella, C. (1973) Appunti su un gruppo di anfore della prima, media
e tarda età Imperiale, In Ostia III: Le terme del Nuotatore: scavo
dell’ambiente V et di un saggio dell’area, Studi Miscellanei 21, Roma,
460-633.
Panella, C. (1980) Retroterra, porti e mercati: l’esempio dell’ager Faler-
nus, In J. H. D’Arms – E. C. Kopff (eds), The Seaborne commerce of
ancient Rome, Studies in Archaeology and History, AAR 36, Rome,
251-259.
Panella, C. (1989) Gli scambi nel Mediterraneo occidentale dal IV al
VII secolo dal punto di vista di alcuni “merci”, In Hommes et richesses
dans l’Empire byzantin (IVe-VIIe siècle), Tome I, Paris, 129-141.
Panella, C. (2001) Le anfore di età imperiale del Mediterraneo occiden-
tale, Céramiques hellénistiques et romaines 3, 177-275.
Panella, C. (2011) Roma, il suburbio e l’Italia in età medio e tardo-
repubblicana: cultura materiale, territori, economie, Facta 4, 11-123.
Panella, C. – Sagui, L. (2001) Consumo e produzione a Roma tra tar-
doantico e altomedioevo: le merci, i contesti, In Roma nell’ Alto medio-
evo, Settimane di studio del Centro Italiano di Studi sull’ Alto Medioevo
48, 757-820.
Panella, C. – Sagui, L. – Coletti, F. (2010) Contesti tardoantichi di
Roma. Una rilettura alla luce di nuovi dati, In S. Menchelli – S. San-
toro – M. Pasquinucci – G. Guiducci (eds), LRCW 3. Late Roman
Coarse Wares, Cooking Wares and Amphorae in the Mediterranean.
Archaeology and Archaeometry. Comparison between western and
eastern Mediterranean, BAR Int. series 2185, Oxford, 57-78.
Papadopoulos, J. K. (1991) Roman Amphorae from the Excavations at
Torone, Archeologiki Ephemerides 128, 67-103.
Papadopoulos, J. K. (2001) “Amphora”, in A. Cambitoglo – J. K. Papa-
dopoulos – O. Tudor Jones (eds), Torone I. The excavations of 1975,
1976, and 1978, Athens, 67-103.
Papanikola-Bakirtzi, D. (2010) Ceramics in Late Antique Thessa-
lonike, In L. Salah Nasrallah – Ch. Bakirtzis – S. J. Friesen (eds), From
Roman to Early Christian Thessalonike. Studies in Religion and Archae-
ology, Boston, 263-298.
Pape, W. (1884) Wörterbuch der griechischen Eigennamen, 3. Aufl. bearb.
von G. E. Benseler, Braunschweig.
Parker, A. J. (1977) Lusitanian amphoras, In Méthodes classiques et
méthodes formelles dans l’étude des amphores, CEFR 32, Rome,
33-46.
Paunier, D. (1981) La Céramique Gallo-Romaine de Genève, Genève –
Paris.
Peacock, D. P. S. – Williams, D. F. (1986) Amphorae and the Roman
Economy, London.
Pecci, A. – Salvini, L. – Cirelli, E. – Augenti, A. (2010) Castor oil
BEZECZKY_Text.indd 232
at Classe (Ravenna-Italy): residue analysis of some late Roman ampho-
rae coming from the port, In S. Menchelli – S. Santoro – M. Pasqui-
nucci – G. Guiducci (eds), LRCW 3. Late Roman Coarse Wares,
Cooking Wares and Amphorae in the Mediterranean. Archaeology
and Archaeometry. Comparison between western and eastern Medi-
terranean, BAR Int. series 2185, Oxford, 617-622.
Pellicioni, M. T. (2012) La Pansiana in Adriatico. Tegole romane per
navigare tra le sponde, Portomaggiore.
Pesavento Mattioli, S. (ed) (1992) Anfore romane a Padova: ritrova-
menti dalla città, Materiali d’archeologia 1, Modena.
Pesavento Mattioli, S. (2000) Anfore betiche in Italia settentrio-
nale: direttrici di approvvigionamento e rapporti con le produzioni
locali (I e II secolo d.C.), In D. Bernal – E. García Vargas (eds), Ex
Baetica Amphorae. Conservas, aceite y vino de la Bética en el Imperio
Romano III, Écija, 733-775.
Pesavento Mattioli, S. (2005) Le anfore da allume. L’apporto di
Padova. Bilancio e prospettive, In P. Borgard – J. P. Brun – M. Picon
(eds), L’ Alun de Méditerranée, Naples – Aix-en-Provence, 177-185.
Pesavento Mattioli, S. (2011) Le anfore Schörgendorfer 558 e il
commercio delle olive adriatiche / Amfore Schörgendorfer 558 i
trigovina jadranskim maslinama, In G. Lipovac Vrkljan – I. Radić
Rossi – B. Šiljeg (eds), Rimske keramičarke I staklarske radonice. Proiz-
vodnja I trgovina na Jadranskom prostoru. Zbornik I. Medunarodnog
Archeološkog Kolokvija, Crikvenica, 23.-24. Listopada 2008, Crik­
venica, 165-173.
Pesavento Mattioli, S. – Carre, M.-B. (2009) Olio e pesce in epoca
romana. Produzione e commercio nelle regioni dell’Alto Adriatico.
Atti del convegno (Padova, 16 febbraio 2007), Antenor Quaderni 15,
Roma.
Pesavento Mattioli, S. – Mazzocchin, S. – Pavoni, M. G. (2000)
I ritrovamenti di anfore presso l’anfiteatro romano di Padova, Bollet-
tino del Museo Civico di Padova 88, 7-44.
Pesavento Mattioli, S. – Zanini, S. (1993) Per un aggiornamento
dell’epigrafia anforaria pataviana. Le Lamboglia 2 e le Dressel 6A del
Museo Civico Archeologico, Bollettino del Museo Civico di Padova 83,
23-60.
Pešić, M. (2013) Connections between Eastern Adriatic coast and
African Provinces in the period from 2nd-5th century AD based on
the underwater finds of amphorae in Croatia, In L. Bombardieri –
A. D’Agostino – G. Guarducci – V. Orsi – S. Valentini (eds), SOMA
2012, Identity and Connectivity, BAR Int. series 2581, Oxford, 1207-
1215.
Petru, S. (1972) Emonske Nekropole (odkrite med leti 1635-1960),
Katalogi in Monografije 7, Izdaja Narodni Muzej v Ljubljani, Lju-
bljana.
Pianetti, F. – Modrzewska-Pianetti, I. (1994) Anfore spagnole del
deposito di Verona (Italia). Interpretazione dei dati analitici, Archivo
Español de Arqueología 67, 147-155.
Piccottini, G. (1997) Amphorae litteratae vom Magdalensberg, In
G. Erath – M. Lehner – G. Schwarz (eds), Komos. Festschrift für Thuri
Lorenz zum 65. Geburtstag, Wien, 203-206.
Piccottini, G. (2000-01) Neues zum Wein- und Lebensmittelimport
in die Stadt auf dem Magdalensberg, Kärnten, Archaeologia Austriaca
84-85, 373-385.
Pieri, D. (1998) Le puits de la rue du Bon-Jésus (îlot 39N). Les amphores,
In M. Bonifay – M.-B. Carre – Y. Rigoir (eds), Fouilles à Marseille. Les
mobiliers (Ier-VIIe s.), Etudes Massaliètes 5, Paris, 231-242.
Pieri, D. (2005) Le commerce du vin oriental à l’époque byzantine (Ve-
VIIe siècles). Le témoignage des amphores en Gaule, Bibliothèque
archéologique et historique 174, Beirut.
Pizzirani, C. (2013) I commerci greci e gli empori del delta padano, In
18.03.19 14:40
 BIBLIOGRAPHY
F. Boschi (ed), Ravenna e l’Adriatico dalle origini all’età romana, Bolo-
gna, 21-40.
Ponsich, M. (1974) Implantation rurale antique sur le Bas-Guadalqui-
vir I, Publications de la Casa de Velazquez, sér. Archéologie II, Madrid.
Porqueddu, A. – Giarrusso, C. – Spanu, P. G. (forthcoming) Le
anfore lusitane del relitto di Punta Sardegna A (Palau, Sardegna). Una
prima caratterizzazione delle forme e degli impasti, In Congreso Interna-
cional. Ãnforas lusitanas. Produçao e difusão, Troia 10-13 outubro 2013.
Portale, E. C. – Romeo, I. (2000) Le anfore locali di Gortina ellenistica
e romana, ReiCretActa 36, 417-426.
Portale, E. C. – Romeo, I. (2000) Contenitori da trasporto, In A. di
Vita (ed), Gortina V.3. Lo Scavo del Pretorio (1989-1995). I Materiali,
Padova, 260-410.
Poulter, A. (2013) Goths on the lower Danube: their impact upon and
behind the frontier / Les Goths du Bas Danube: leur impact sur et au-
delà de la frontière, Antiquité tardive 21, 63-76.
Pröttel, P. M. (1996) Mediterrane Feinkeramik des 2.-7. Jahrhunderts
n. Chr. im oberen Adriaraum und in Slowenien, Kölner Studien zur
Archäologie der Römischen Provinzen 2, Espelkamp.
Quiri, E. (2009) Importazioni di anfore altoadriatiche a Torino, In
S. Pesavento Mattioli – M.-B. Carre (eds), Olio e pesce in epoca
romana. Produzione e commercio nelle regioni dell’Alto Adriatico.
Atti del convegno (Padova, 16 febbraio 2007), Antenor Quaderni 15,
Roma, 293-300.
Quiri, E. (2011) Le anfore: un esempio di reimpiego, In L. Brecciaroli
Taborelli (ed), Oro, pane e scrittura, Studi e ricerche sulla Gallia Cis-
alpina 24, 107-116.
Quiri, E. – Spagnolo Garzoli, G. (2015) Imports of alum from
Milos to Novara (Italy), In S. Demesticha (ed), Per Terram, per Mare.
Seaborne Trade and the Distribution of Roman Amphorae in the Medi-
terranean, Uppsala, 181-188.
Radić, I. – Jurišić, M. (1993) Das antike Schiffswrack von Mljet, Kroa-
tien, Germania 71, 113-138.
Radić-Rossi, I. (2007) Podmorsko arheolosko nalaziste u predjelu spi-
nut u splitu, Bastina 34, 351-372.
Rambla Torralvo, J. A. – Mayorga Mayorga, J. (2007) Hornos
de época altoimperial en calle Carretería, Málaga, In Figlinae Malaci-
tanae. La producción de cerámica romana en los territorios malacitanos,
Malaga, 61-78.
Rauh, N. K. – Slane, K. W. (2000) Possible amphora kiln sites in
W Rough Cilicia, JRA 13, 319-330.
Remesal Rodríguez, J. (1977-78) La economía oleícola bética: nuevas
formas de análisis, Archivo Español de Arqueología 50-51, 87-142.
Remesal Rodríguez, J. (1996) Mummius Secundinus. El Kalendarium
Vegetianum y las confiscaciones de Severo en la Bética, HA Severus
12-13, Gerión 14, 195-221.
Remesal Rodríguez, J. (1998) Baetican olive oil and the Roman
economy, In S. Keay (ed), The Archaeology of early Roman Baetica,
JRA Suppl. 29, Portsmouth – Rhode Island, 183-199.
Remesal Rodríguez, J. – García Sánchez, M. (2007) Los tituli
picti sobre ánforas olearias orientales, In J. M. Blázquez Martínez –
J. Remesal Rodríguez (eds), Estudios sobre el Monte Testaccio (Roma)
IV, Col·lecció Instrumenta 24, Barcelona, 173-182.
Remollà Vallverdú, J. (2000) Las ánforas tardo-antiguas en Tarraco
(Hispania Tarraconensis). Siglos IV-VII d.C., Col·lecció Instrumenta 7,
Barcelona.
Rémy, B. – Jospin, J. P. (1998) Trois graffites sur céramique du musée
d’Aoste (Isère), Zeitschrift für Papyrologie und Epigraphik 122, 263-
264.
Reynolds, P. (1993) Settlement and pottery in the Vinalopó Valley (Ali-
cante, Spain): A.D. 400-700, BAR Int. series 588, Oxford.
BEZECZKY_Text.indd 233
233
Reynolds, P. (2000) Baetican, Lusitanian and Tarraconensian amphorae
in classical Beirut: some preliminary observations of trends in amphora
imports from the western Mediterranean in the Anglo-Lebanese exca-
vations in Beirut (BEY 006, 007 and 045), In D. Bernal – E. García
Vargas (eds), Ex Baetica Amphorae. Conservas, aceite y vino de la Bética
en el Imperio Romano III, Écija, 1035-1060.
Reynolds, P. (2004) The Roman pottery from the Triconch palace, In
R. Hodges – W. Bowden – K. Lako, Byzantine Butrint: Excavations
and surveys 1994-99, Oxford, 224-269.
Reynolds, P. (2005) Levantine Amphorae from Cilicia to Gaza: A
typology and analysis of regional production trends from the 1st to 7th
centuries, In J. M. Gurt i Esparraguera – J. Buxeda – M. A. Cau (eds),
LRCW 1. Late Roman Coarse Wares, Cooking Wares and Amphorae
in the Mediterranean, BAR Int. series 1340, Oxford, 563-611.
Reynolds, P. (2010) Hispania and the Roman Mediterranean, AD 100-
700. Ceramics and Trade, London.
RIB; Collingwood, R. G. – Wright, R. P. (1994) The Roman Inscrip-
tions of Britain II, 6, Stroud.
Righini, V. (1998) I bolli laterizi di età romana nella Cispadana. Le
Figlinae. Prima/seconda parte, In V. Righini (ed), Le fornaci romane.
Produzione di anfore e laterizi con marchi di fabbrica nella Cispadana
Orientale e nell’Alto Adriatico. Atti delle Giornate Internazionali di Stu-
dio (Rimini, 16-17 Ottobre 1993), Rimini, 29-67.
Righini, V. – Biordi, M. – Pellicioni, M. T. (1993) I bolli laterizi
romani della regione Cispadana (Emilia e Romagna), In C. Zaccaria
(ed), I laterizi di età romana nell’area nordadriatica, Rome, 23-91.
Riley, J. A. (1979) The coarse pottery from Benghazi. Sidi Khrebish
Excavations, Benghazi (Berenice) II, Supplement to Libya Antiqua 5,
Tripoli, 91-497.
Riva, F. (1987) Anfore romane de Chieri, In G. Cresci Marrone (ed),
Museo Archeologico di Chieri: contributi alla conoscenza del territorio in
età romana, Torino, 90-115.
Rizzo, G. (2003) Instrumenta urbis I. Ceramiche fini da mensa, lucerne
ed anfore a Roma nei primi due secoli dell’Impero, CEFR 307, Roma.
Rizzo, G. (2014) Le anfore, Ostia e i commerci Mediterranei, Ostia VI,
Le Terme del Nuotatore, Studi Miscellanei 38, 73-370.
Rizzo, G. – Capone, M. – Constantini, C. – Gafà, R. – Penti­
ricci, M. – Munzi, M. (2004) Vigna Barberini, settore D. Periodo
IV: 540/550-580/590 d. C., In L. Paroli – L. Vendittelli (eds), Roma
dall’antichità al medioevo II, contesti tardoantichi e altomedievali,
Milan.
Rodríguez Almeida, E. (1981) Varia de Monte Testaceo, CuadRom
15, 105-164.
Rodríguez Almeida, E. (1984) Il Monte Testaccio: ambiente, storia,
materiali, Rome.
Rodríguez Almeida, E. (1991) Su alcuni curiosi graffiti anforari dal
Monte Testaccio, Bullettino della Commissione Archeologica Comunale
di Roma 93-1, 35-40.
Rodríguez Almeida, E. (1993) Grafitti e produzione anforaria della
Betica, In W. V. Harris (ed), The Inscribed Economy. Production and
Distribution in the Roman Empire in the Light of Instrumentum
Domesticum, JRA Suppl. 6, Rome, 95-107.
Rodríguez Almeida, E. (2000) El Monte Testaccio, ayer, hoy y
mañana, In Ex Baetica Amphorae. Conservas, aceite y vino de la Bética
en el Imperio Romano (Congreso Internacional, Sevilla-Écija, 1998),
Écija, 1279-1302.
Rostovtzeff, M. (1957) The Social and Economic History of the
Roman Empire, Oxford.
Rouquette, D. (1980) Nouvelles marques sur amphores à huile de
l’Hérault, Archéologie en Languedoc 3, 151-160.
Rudolph, W. W. (1979) Excavations at Porto Cheli and Vicinity. Prelimi-
18.03.19 14:40
 234 BIBLIOGRAPHY
nary Report V: The Early Byzantine Remains, Hesperia 48, 294-320.
Sáez Romero, A. M. (2010) Tradizione Fenicia versus Romanizzazione.
Le anfore di Gadir/Gades in epoca ellenistica e i suoi centri produttori,
ReiCretActa 41, 509-522.
Saguì, L. (1998) Il deposito della Crypta Balbi: una testimonianza
imprevedibile sulla Roma del VII secolo?, In L. Saguì (ed), Ceramica
in Italia: VI-VII secolo, Florence, 305-330.
Sakl-Oberthaler, S. (2000) Die römischen Transportamphoren aus
den Altgrabungen in Flavia Solva, FÖ 39, 381-424.
Salama, P. (2006) Sites commerciaux antiques sur le littoral de l’Algérois,
MEFRA 118-2, 527-547.
Šašel, J. – Petru, P. (1971) Claustra Alpium Iuliarum 1, Katalogi in
monografije Narodni muzej v Ljubljani 5, Limes v Jugoslaviji 2, Beo-
grad.
Sauer, R. (2005) Ergebnisse der mineralogisch-petrographischen Ana-
lysen von ausgewählten römischen Amphorenproben aus Wien, In
F. Krinzinger (ed), Vindobona, AForsch 12, DenkschrWien 328, Wien,
109-142.
Sauer, R. (2013) Petrology, In T. Bezeczky, The Amphorae of Roman
Ephesus, FiE XV/1, Wien, 197-212.
Schimmer, F. (2009) Amphoren aus Cambodunum (Kempten). Ein
Beitrag zur Handelsgeschichte der römischen Provinz Raetia, Münch-
ner Beiträge zur Provinzialrömischen Archäologie 1, Cambodunumfor-
schungen 7, Wiesbaden.
Schindler-Kaudelka, E. (2000) Un lot d’amphores tibériennes tar-
dives. AA44, la cave à provisions de la fabrica impériale du Magdalens-
berg, In SFECAG, Actes du congrès de Libourne, Marseille, 387-399.
Scholz, M. (2012) Ziegelrechnungen – Aspekte der Organisation römi-
scher Ziegeleien, In M. E. Fuchs – R. Sylvestre – C. Schmidt Heiden-
reich (eds), Inscriptions mineures: nouveautés et réflexions, Actes du
premier colloque Ductus, Lausanne – Bern, 339-357.
Schönberger, H. – Simon, H.-G. (1983) Die Kastelle in Altenstadt,
Limesforschungen 22, Berlin.
Schrunk, I. – Begović, V. (2000) Roman estates on the Island of Bri-
juni, Istria, JRA 13, 253-276.
Schrunk, I. – Begović, V. (2011) Osnivanje novih biskupija u VI. Stol-
jeću na primjeru Brijuna, In Zbornik u čast Emilija Marina za 60.
rođenan, Split, 657-672.
Sciallano, M. – Sibella, P. (1994) Amphores: comment les identifier?,
Aix-en-Provence.
Shkodra, B. (2005) Ceramica e commercio a Durrës: evidenza prelimi-
nare dai contesti del VI secolo nel macellum-forum, Quaderni Friulani
di Archeologia 15-1, 131-154.
Slane, K. (2008) Corinth’s trade with the Adriatic, ReiCretActa 40, 237-
241.
Slapšak, B. (1974) Tegula Q. Clodi Ambrosi, Situla 14-15, 173-181.
Slapšak, B. (1977) AD: CIL 5,698 (Materija): “Via Decreta - Translata
(in fines alicuius) - Restituta”, Arheološki Vestnik 28, 122-128.
Sodini, J. (1984) Aliki 2. La basilique double, Athen – Paris.
Solin, H. (1966-67) Eine Amphoreninschrift vom Monte Testaccio mit
einer neuen Datierungsformel, RM 73-74, 266-269.
Starac, A. (1995) Carski posedi u Histriji, Opusc. Arheol. 18, 133-145.
Starac, A. (1994-95) Morfologija Sjevernojaranskih amfora: Primerji iz
Istre, Diadora 16-17, 135-162.
Starac, A. (2001) Produzione e distribuzione delle anfore nord-adria­
tiche nell’Istria, ReiCretActa 31, 269-277.
Starac, A. (2006) Promet amforama prema nalazima u Rovinjskome
podmorju, Histria Archaeologica 37, 85-116.
Starac, A. (2008) Nalaz rimskog svetišta u četvrti sv. Teodora u Puli.
Arheološka istraživanja 2008 / Discovery of Roman Sanctuary in the
Quarter of St. Theodore in Pula. Archaeological Campaign of 2008,
Histria Archaeologica 38-39, 123-168.
BEZECZKY_Text.indd 234
Starac, A. (2010a) The workshop of Laecanius at Fažana. Some recent
testimonies, ReiCretActa 41, 61-65.
Starac, A. (ed) (2010b) Dragonera. Dva bsera / Dragonera. Two pearls,
Monografije i Katalozi 19, Archeoloski Muzej Istre, Pula.
Starac, A. – Matijašić, R. (1991) Skupni nalaz amfora u Puli 1991.
godine / A hoard of amphorae discovered in Pula in 1991, Diadora
13, 77-101.
Stoppioni Piccoli, M. L. (1983) Le anfore, In G. Bermond Montanari
(ed), Ravenna e il porto di Classe: Venti anni di ricerche archeologiche tra
Ravenna e Classe, Imola, 130-146.
Stoppioni, M. L. (1993) Le anfore, In M. L. Stoppioni (ed), Con la terra
e con il fuoco. Fornaci romane del Riminese, Rimini, 145-154.
Strazzulla Rusconi, M. J. (1984) Bolli laterizi, In Problemi storici
ed archeologici dell’Italia nord-orientale e delle regioni limitrofe
della preistoria al medioevo, AttiMusTrieste 12, 2, Trieste, 135-155.
Suić, M. (1987) Cissa Pullaria, Baphium Cissense, Episcopus Cessense,
Arheološki radovi i rasprave 10, 185-219.
Sztetyłło, S. – Borkowsky, Z. (1986) Un timbre céramique byzantin
de l’éparque de Constantinople, In J. Y. Empereur – Y. Garlan (eds),
Recherches sur les amphores grecques, BCH Suppl. 13, Paris, 649-654.
Taeuber, H. (2011) Der Gaius Laecanius Bassus-Brunnen in Ephe-
sos – privat oder öffentlich finanziert?, NFN Imperium and Officium.
Comparative Studies in Ancient Bureaucracy and Officialdom, Working
Papers (IOWP) Version 01, May 2011, http://iowp.univie.ac.at/sites/
default/files/CLB_Inschr_%20IOWP_03.pdf.
Tassaux, F. (1982) Laecanii, Recherches sur une famille sénatoriale
d’Istrie, MEFRA 94, 227-269.
Tassaux, F. (1983-84) L’implantation territoriale des grandes familles
d’Istrie sous le Haut-Empire romain, AttiMusTrieste 13, 2, 193-229.
Tassaux, F. (1990) Assises économiques des aristocraties et “bourgeoi-
sies” des cités istriennes sous le Haut-Empire romain, In La Città
nell’Italia settentrionale in età romana. Morfologia, strutture e funzio-
namento dei centri urbani delle Regiones X e XI. Atti del convegno di
Trieste, 13-15 marzo 1987, Rome, 69-128.
Tassaux, F. (1998) Apports récents de l’épigraphie à l’histoire écono-
mique et sociale de Brioni, In G. Paci (ed), Epigrafia romana in area
adriatica. IXe Recontre franco-italienne sur l’épigraphie du monde
romain, Macerata 10-11 Nov. 1995, Pisa – Roma, 77-99.
Tassaux, F. (2001) Production et diffusion des amphores à huile
istriennes, Antichità Altoadriatiche 46, 501-543.
Tassaux, F. (2005) Patrimoines sénatoriaux de la Decima Regio, Cahiers
du Centre Gustave Glotz 16, 139-164.
Tassaux, F. (2010) Les milliardaires de l’Adriatique romaine, Exhibition
catalogue 19 May to 18 June 2010, Bordeaux.
Tassaux, F. – Matijašić, R. – Kovačić, V. (2001) Loron (Croatie).
Un grand centre de production d’amphores à huile istriennes (Ier-IVe s.
p. C.), Mémoires / Ausonius 6, Bordeaux.
Taylor, M. V. – Collingwood, R. G. (1925) Roman Britain in 1925,
JRS 15, 223-252.
Tchernia, A. (1986) Le vin de l’Italie romaine, BEFAR 261, Rome.
Tchernia, A. – Zevi, F. (1972) Amphores vinaires de Campanie et de
Tarraconaise à Ostie, In Recherches sur les amphores romaines, CEFR
10, Rome, 35-76.
Testaccio 1999; Casulleras, J. – Garcia Brosa, G. – Garcia
Morcillo, M. – Rovira Guardiola, R., 4. Los grafitos del
siglo II (Campañas de 1989 y 1990), In J. M. Blázquez Martínez –
J. Remesal Rodríguez (eds), Estudios sobre el Monte Testaccio
(Roma) I, Col·lecció Instrumenta 6, Barcelona, 53-73.
Testaccio 2001; Remesal, J. – Aguilera, A., 2. Los tituli picti, 3. Los
sellos, In J. M. Blázquez Martínez – J. Remesal Rodríguez (eds), Estu-
dios sobre el Monte Testaccio (Roma) II, Col·lecció Instrumenta 10,
Barcelona, 45-203. 205-263.
18.03.19 14:40
 BIBLIOGRAPHY
Testaccio 2003; Remesal, J. – Rovira Guardiola, R. – García
Brosa, G. – Ozcáriz Gilet, P., 4. Los grafitos, In J. M. Blázquez
Martínez – J. Remesal Rodrígue (eds), Estudios sobre el Monte Testac-
cio (Roma) III, Col·lecció Instrumenta 14, Barcelona, 363-397.
Testaccio 2010a; Remesal, J., 3. Los sellos, In J. M. Blázquez Martínez –
J. Remesal Rodríguez (eds), Estudios sobre el Monte Testaccio (Roma)
V, Col·leció Instrumenta 35, Barcelona, 167-241.
Testaccio 2010b; Remesal, J. – Calzada Baños, S. – Rovira Guar-
diola, R.– Soria Rincón, J., 4. Los grafitos, In J. M. Blázquez Mar-
tínez – J. Remesal Rodríguez (eds), Estudios sobre el Monte Testaccio
(Roma) V, Col·lecció Instrumenta 35, Barcelona, 243-371.
Tiussi, C. (2007) Importazione vinaria ad Aquileia in età repubblicana:
Le anfore rodie, In G. Cuscito – C. Zaccaria (eds), Aquileia dalle
origini alla costituzione del ducato longobardo: territorio, economia,
società. Atti della 37 settimana di studi aquileiesi, 18-20 maggio 2006,
Antichità Altoadriatiche 65, Trieste, 479-496.
Tiussi, C. (2008) Le anfore, In F. Maselli Scotti (ed), Luoghi di vita
rurale. Un percorso che attraversa i secoli, Monfalcone, 102-105.
Tiussi, C. (2010) Le anfore dagli scavi della chiesa di San Michele, Nota
preliminare, In Michele, il guerriero celeste. L’abbazia di San Michele
di Cervignano del Friuli: la storia, lo scavo, il culto, Aquileia, 266-279.
Tomber, R. (2005) Trade relations in the eastern Mediterranean and
beyond. The Egyptian-Indian connection, In M. Berg Briese – L. Erik
Vaag (eds), Trade relations in the Eastern Mediterranean from the Late
Hellenistic period to Late Antiquity. The Ceramic evidence, Halicar-
nassian Studies 3, Odense, 221-233.
Tomber, R. (2008) Indo-Roman Trade. From pots to pepper, London.
Tomlin, R. S. O. (2009) III. Inscriptions, Britannia 40, 313-363.
Toniolo, A. (1991) Le Anfore di Altino, Monografia di Archeologia
Veneta 14, Padua.
Toniolo, A. (2011) L’inedita attività di un capostipite / Neobjavljena
aktivnost jednoga pretka, In G. Lipovac Vrkljan – I. Radić Rossi –
B. Šiljeg (eds), Rimske keramičarke I staklarske radonice. Proizvod-
nja I trgovina na Jadranskom prostoru. Zbornik I. Medunarodnog
Archeološkog Kolokvija, Crikvenica, 23.-24. Listopada 2008, Crik­
venica, 189-192.
Ugarković, M. (2016) Light from the East. Hellenistic relief lamps
of eastern Mediterranean provenance in Dalmatia, In G. Lipovac
Vrkljan – I. Ožanić – M. Ugarković (eds), Roman and Late Antique
Lamps: Production, Distribution, Contacts on the Mediterranean, Zag-
reb, 5-18.
Ujčić, Ž. (2005) Das frühchristliche Istrien, In K. Džin et al. (eds),
Histria. Istra. Istria. Ein Archäologisches Juwel in der Adria, Mono-
grapfije i Katalozi. Arheološki Muzej Istre 15, Zagreb, 87-104.
van der Werff, J. H. (1986) The Amphora Wall in the House of the
Porch, Ostia. Excavations in the House of the Porch (V.II.4-5) at
Ostia, Bulletin antieke beschaving 61, 96-137.
van Hommerich, L. (1966) Nieuws van de Westelijke Romeins-Heer-
lense woonsektor, Het Land van Herle 16, 6-12.
Vaz Pinto, I. – De Almeida, R. R. – Martin, A. (eds) (2016) Lusita-
nian Amphorae: Production and Distribution, Roman and Late Anti-
que Mediterranean Pottery 10, Oxford.
Vazzoler, L. (2009) Laterizi con marchio TROSI provenienti dallo
scavo della Motta di Foghini, Società Friulana di Archeologia 23, 149-
151.
Véninger, P. (2016) Flis és terra rossa fazekas szemmel. Koron-
gozhatósági vizsgálat: milyen agyag lehet alkalmas nagyméretű római
amphorák készítésére? / Flysch and terra rossa from potter’s point
of view. Studiing by potter’s wheel: what kind of clay might be used
for large-sized Roman amphora production?, Archeometriai Műhely,
www.ace.hu/am 2016. XIII. 2, 131-140.
BEZECZKY_Text.indd 235
235
Ventura, P. – Degrassi, V. – Riavez, P. (2014) Finds from Late
Roman and Early Medieval Tergeste (I), In N. Poulou-Papadimi-
triou – E. Nodarou – V. Kilikoglou (eds), LRCW 4. Late Roman
Coarse Wares, Cooking Wares and Amphorae in the Mediterranean.
Archaeology and Archaeometry. The Mediterranean: A market with-
out frontiers, BAR Int. series 2616, Oxford, 955-961.
Verzár-Bass, M. (ed) (1991) Scavi ad Aquileia 1,1. L’area a est del Foro.
Rapporto degli scavi 1988, Studi e ricerche sulla Gallia Cisalpina 3,
Roma.
Verzár-Bass, M. (ed) (1994) Scavi ad Aquileia 1,2. L’area a est del Foro.
Rapporto degli scavi 1989-91, Studi e ricerche sulla Gallia Cisalpina
6, Roma.
Vežić, P. (2005) Zadar na pragu kršćanstva, Arhitektura ranoga kršćan-
stva u Zadru i na zadarskome području, Zadar, 38-41.
Vicelja Matijašić, M. (2007) Istra i Bizant. Neki povijesno-ikonografski
aspekti u interpretaciji umjetnosti 6. stoljeća u Istri, Rijeka.
Vidošević, I. (2003) Rimska keramika s lokaliteta Stračevićeve ulice 37
u Sisku, Godišnjak Gradskog Muzeja Sisak II-IV, Sisak, 11-74.
Vidrih Perko, V. (1997) Some late Roman ceramic finds from the Slo-
venian Karst region, ReiCretActa 35, 249-258.
Vidrih Perko, V. (2006) Amphorae of the Eastern Adriatic and its
North­eastern Hinterland, In S. Čače – A. Kurilić – F. Tassaux (eds),
Les roues de l’Adriatique antique. Géographie et économie / Putovi
antičkog Jadrana. Geografija i gospodarstvo, Bordeaux – Zadar, 209-
224.
Vidrih Perko, V. – Pavletić, M. (2000) Report on ceramics from Bri-
juni, ReiCretActa 36, 263-269.
Vidrih Perko, V. – Župančić, M. (2005) Amphorae in Western Slove-
nia and in Northern Istria, In J. M. Gurt i Esparraguera – J. Buxeda –
M. A. Cau (eds), LRCW 1. Late Roman Coarse Wares, Cooking Wares
and Amphorae in the Mediterranean, BAR Int. series 1340, Oxford,
521-536.
Vidrih Perko, V. – Župančić, M. (2011) Local brick and ampho-
rae production in Western Slovenia / Lokalna provizvodnja opeke I
amfora u zapadnoj Sloveniji, In G. Lipovac Vrkljan – I. Radić Rossi –
B. Šiljeg (eds), Rimske keramičarke I staklarske radonice. Proizvod-
nja I trgovina na Jadranskom prostoru. Zbornik I. Medunarodnog
Archeološkog Kolokvija, Crikvenica, 23.-24. Listopada 2008, Crik­
venica, 151-162.
Villa, L. (1998) Alcuni aspetti della circolazione di prodotti di impor-
tazione in Friuli tra VI e VII secolo, In L. Saguì (ed), Ceramica in Ita-
lia: VI-VII secolo, Florence, 275-288.
Villa, L. (2013) Le anfore, In C. Tiussi – L. Villa – M. Novello (eds),
Constantino e Teodoro. Aquileia nel IV secolo, Milan, 240-244.
Višnjić, J. (2009) Amfore, In L. Bekic – N. Radic-Stivic (eds), Tarsati-
cki Principij. Kasnoanticko vojno zapovjednistvo. Principia di Tarsatica.
Quartiere generale d’epoca tardoantica, Rijeka, 123-142.
Višnjić, J. – Bekić, L. – Pleština, I. (2010) Arheoloska istrazivanja na
prostoru anticke uljare u uvali Maric (Porto Marricio) kod Barbarige,
Portal, Godisnjak Hrvatskog restauratorskog zavoda 1, 229-260.
Vitasović, A. (2003) Crkva Sv. Marije kasnoantička crkva 5. st. - bizan-
tinska bazilika 6. st. / Church of St. Mary - Late Roman church, 5th cen-
tury - Byzantine basilica, 6th century, Histria Archaeologica 34, 61-101.
Vitasović, A. (2004) Istraživanje i zaštita rimske vunare kao zasebnog
segmenta ladanjskog dvorca na tri terase u zaljevu Verige na otoku
Veliki Brijun, Histria Archaeologica 35, 89-110.
Vitasović, A. (2005) Antički objecti uvali Dobrinka na otoku Veliki
Brijun / Antique facilities in Dobrinka cove on the island of Veliki
Brijun, Histria Archaeologica 36, 157-210.
Vitasović, A. (2007) Antique facilities in Dobrika Cove on the island of
Veliki Brijun, Histria Archaeologica 36, 157-210.
18.03.19 14:40
 236 BIBLIOGRAPHY
Volpe, G. (1990) La Daunia nell’età della romanizzazione. Paesaggio
agrario, produzione, scambi, Bari.
Volpe, G. – Annese, C. – Disantarosa, G. – Leone, D. (2010) Pro-
duzioni locali ed importazioni nella Puglia centro-settentionale, In
N. Poulou-Papadimitriou – E. Nodarou – V. Kilikoglou (eds), LRCW
4. Late Roman Coarse Wares, Cooking Wares and Amphorae in the
Mediterranean. Archaeology and Archaeometry. The Mediterranean:
A market without frontiers, BAR Int. series 2616, Oxford, 643-656.
Warner Slane, K. – Sanders, G. D. R. (2005) Corinth: Late Roman
Horizons, Hesperia 74, 243-279.
Weisshäupl, R. (1899) Funde in Pola und Umgebung, ÖJh 2, 77-82.
Weisshäupl, R. (1900) Funde in Südistrien, ÖJh 3, 193-204.
Weiss-König, S. (2011) Graffiti auf römischer Gefäßkeramik aus dem
Bereich der Colonia Ulpia Traiana/Xanten, Xantener Berichte 17,
Xanten.
Weszburg, T. – Papp, G. (1987) X-ray powder diffraction analyses, In
T. Bezeczky (ed), Roman Amphorae from the Amber Route in West-
ern Pannonia, BAR Int. series 386, Oxford, 128-133.
Will, E. L. (1970) Les Timberes Amphoriques Latins, Délos 27, Paris,
383-386.
Will, E. L. (1983) Exportation of olive oil from Baetica to the Eastern
Mediterranean, In Producción y comercio del aceite en la Antigüedad.
Segundo Congreso Internacional, Madrid, 391-440.
Will, E. L. (1987) The Roman amphoras, In A. M. McCann – J. Bour-
geois – E. Gazda – J. P. Oleson – E. L. Will (eds), The Roman Port and
Fishery of Cosa. A Center of Ancient Trade, Princeton – New Jersey,
171-220.
Will, E. L. (1989a) Relazioni mutue tra le anfore romane. I ritrovamenti
in Oriente alla luce dei dati ottenuti nell’Occidente, In Amphores
romaines et histoire économique. Dix ans de recherches, Actes du col-
loque de Sienne (22-24 mai 1986), CEFR 114, Rome, 297-309.
Will, E. L. (1989b) Roman Amphoras, In S. MacNally – I. V. Schrunk
(eds), Diocletians Palace 6, Minneapolis, 58-113.
Will, E. L. (1996) Mediterranean shipping amphoras from the 1941-50
excavations, In V. Begley et al. (eds), The Ancient Port of Arikamedu.
New Excavations and Researches 1989-92, Mémoires archéologiques
22, Paris, 317-349.
Will, E. L. (1997) Shipping Amphoras as Indicators of Economic
Romanization in Athens, In M. C. Hoff – S. I. Rotroff (eds), The
Romanization of Athens. Proceedings of an International Confer-
ence held at Lincoln, Nebraska, April 1996, Oxbow Monographs 94,
Oxford, 117-133.
Williams, D. – Lund, J. (2013) Petrological Analyses of “Pinched-
handle” Amphorae from the Akamas Peninsula, Western Cyprus,
In M. Lawal – J. Lund (eds), The Transport Amphorae and Trade of
Cyprus, Gösta enbom Monographs 3, Aarhus, 155-164.
Williams, D. – Panella, C. – Keay, S. – Carreras, C. – Bonifay,
M. (2005a) Dressel 2-4, In S. Keay – D. F. Williams, Roman Ampho-
rae. A digital resource, University of Southampton, http://archaeology­
data­service.ac.uk/archives/view/amphora_ahrb_2005.
Williams, D. F. – Panella, C. – Rizzo, G. (2005b) Forlimpopoli
type, In S. Keay – D. F. Williams (eds), Roman Amphorae. A digital
resource, University of Southampton, http://archaeologydataservice.
ac.uk/archives/view/amphora_ahrb_2005.
BEZECZKY_Text.indd 236
Woodworth, J. – Bernal, D. – Bonifay, M. – De Vos, D. – Gar-
nier, N. – Keay, S. – Pecci, A. – Poblome, J. – Pollard, M. –
Richez, F. – Wilson, A. (2015) The content of African Keay 25/
Africana 3 amphorae: initial results of the CORONAM project, In
C. Oliveira – R. Morais – A. Morillo Cerdán (eds), ArchaeoAnalytics.
Chromatocraphy and DNA Analysis in Archaeology, Esposende, 41-57.
Wright, R. P. (1960) Roman Britain in 1959: I. Sites Explored.
II. Inscriptions, JRS 50, 210-242.
Wright, R. P. – Wilson, D. R. (1963) Roman Britain in 1962: I. Sites
Explored. II. Inscriptions, JRS 53, 125-167.
Zaccaria, C. (1989) Per una prosopografia dei personaggi menzionati
sui bolli delle anfore romane, dell’Italia nordorientale, In Amphores
romaines et histoire économique dix ans de recherche, CEFR 114,
469-488.
Zaccaria, C. (1991) La ricerca sull’instrumentum inscriptum nell’Italia
nordorientale: esperienze e problemi, In Specimina Nova Universitatis
Quinqueecclesiensis VII-1, Pécs, 301-323.
Zaccaria, C. – Gomezel, C. (2000) Aspetti della produzione e cir-
colazione dei laterizi nell’area adriatica settentrionale tra II secolo
a.C. e II secolo d.C., In P. Boucheron – H. Broise – Y. Thébert (eds),
La Brique antique et médiévale. Production et commercialisation d’un
matériau. Actes Colloque International, Saint-Cloud, 16-18 novembre
1995, Rome, 285-310.
Zaccaria, C. – Župančić, M. (1993) I bolli laterizi del territorio di
Tergeste romana, In C. Zaccaria (ed), I laterizi di età romana nell’area
nordadriatica, Rome, 135-178.
Žerjal, T. (2008) Trade and consumption of foodstuffs in Northern
Adriatic: the case of the villa Školarice, In R. Auriemma – S. Karinja
(eds), Terre di Mare, Atti del Convegno Internazionale di Studi, Trieste,
8-10 novembre 2007, Trieste – Pirano, 462-469.
Žerjal, T. (2010) Skolarice near Koper-Some Late Roman context in
the norther Adriatic, In S. Menchelli – S. Santoro – M. Pasquinucci –
G. Guiducci (eds), LRCW 3. Late Roman Coarse Wares, Cooking
Wares and Amphorae in the Mediterranean. Archaeology and Archae-
ometry. Comparison between western and eastern Mediterranean,
BAR Int. series 2185, Oxford, 703-710.
Zevi, F. (1966) Appunti sulle anfore romane. La tavola tipologica del
Dressel, Archaeologia Classica 18, 208-247.
Zevi, F. (1967) Anfore istriane ad Ostia. Nota sul commercio istriano,
AttiMemIstria 67, 21-33.
Živkovic, S. – Boger, D. (2006) Coprolite status of Coptocampylodon
lineolatus Elliott 1963 (incertae sedis) from Middle Eocene deep-sea
sediments of Istria (Croatia), Micropaleontology 52, 371-379.
Zmaić, V. (2013) Late Byzantine amphorae from eastern Adriatic under-
water sites, Skyllis. Zeitschrift für Unterwasserarchäologie 13, 81-88.
Zollt, T. (1998) Das ionische Kämpferkapitell. Definitionsprobleme,
In U. Peschlow – S. Möllers (eds), Spätantike und byzantinische
Bauskulptur. Beiträge eines Symposions in Mainz, Februar 1994,
Forschungen zur Kunstgeschichte und christlichen Archäologie 19,
Stuttgart, 62.
Župančić, N. – Horvat, J. – Bolc, M. (1998) The Production of
Graeco-Italic Amphorae in the Adriatic Region, Materiali in Geokolje
45, 3-4, 345-357.
18.03.19 14:40
 237

CONTRIBUTOR S

TAMÁS BEZECZKY (†) HORACIO GONZÁLEZ CESTEROS

Senior Research Fellow Austrian Archaeological Institute
Institute for the Study of Ancient Culture Franz-Klein-Gasse 1, 1190 Vienna, Austria
Hollandstrasse 11-13, 1020 Vienna, Austria horacio.gonzalez@oeai.at

PIERO BERNI MILLET SÁNDOR JÓZSA

UMR5140 „ASM“-Montpellier (équipe TP2C) /
LabEx Archimede ANR-11-LABX-0032-01 /
Catalan Institute of Classical Archaeology (ICAC,
Tarragona)
pbernim@gmail.com; pberni@icac.cat
Department of Petrology and Geochemistry, Institute of
Geography and Earth Sciences, Eötvös Loránd University
Budapest
Pázmány sétány 1/c, Budapest, 1117, Hungary
sandor.jozsa@geology.elte.hu

MICHEL BONIFAY MARTINO LA TORRE

Directeur de Recherche au CNRS, MMSH – Centre
Camille Jullian
5 rue du Château de l’Horloge BP 647, 13094 Aix-en-
Provence, France
mbonifay@mmsh.univ-aix.fr
CLAUDIO CAPELLI
Contracted researcher, Dipartimento di Scienze della Terra,
dell’Ambiente e della Vita (DISTAV), Università degli
Studi di Genova, Italy /
Associate collaborator to the Centre Camille Jullian
(Aix Marseille Université, CNRS, Ministère de la Culture
et de la Communication, INRAP, CCJ, UMR 7299, 13094
Aix-en-Provence, France)
capelli@dipteris.unige.it
Büro für Bauaufnahme und Bauforschung
Fritz-Kalle-Straße 11, 65187 Wiesbaden, Germany
latorre@bfb-wiesbaden.de; www.bfb-wiesbaden.de
ALEXANDER SCHOBERT
alex_schobert@hotmail.com
GYÖRGY SZAKMÁNY
Department of Petrology and Geochemistry, Institute of
Geography and Earth Sciences, Eötvös Loránd University
Budapest
Pázmány sétány 1/c, Budapest, 1117, Hungary
gyorgy.szakmany@geology.elte.hu

BEZECZKY_Text.indd 237 18.03.19 14:40

BEZECZKY_Text.indd 238 18.03.19 14:40