M. Járó - L. Költő szerk.: Archaeometrical research in Hungary (Budapest, 1988)

Analysis - BALLA Márta, BÉRCZI János, KEÖMLEY Gábor, ROSNER Gyula, GABLER Dénes: Provenance studies of ceramics by neutron actiwtion analysis

f 1 26 24 22 8 17 16 20 13 15 14 21 19 10 7 10 2 A 3 5 1. Í GAUL j * NORTH-ÏTALY ! ARRETINE I I t i « I Fig. 1 Dendrogram of Gaulish, North-Italian and Arretinian samples was made in an Arretine workshop. On the basis of this we can state that an export of Arretium spread over the more distant Danube provinces, like Moesia. The aim of the examination of South—Gaulish (58—73) samples was the separation of potsherd from La Graufesenque and Banassac; this separation is of chronological importance. Chemical analysis seemed to be suitable as a means of separation. The two groups polarized in the dendrogram obtained as a result of cluster analysis (Fig. 2). Ten samples (37-46) enabled us to define the Central-Gaulish workshops. The analyses made possible the separation of the plain and undecorated South— and CentralGaulish, Lezoux and Rheinzabern sherds (Fig. 3). According to the analysis of samples from Rheinzabern (27-36, 47, 49-56) it seems that the products of Rheinish manufacture can be separated not only from CentralGaulish or Westerndorf sherds (Fig. 4) but using elemental concentration values the sherds of the Antoninian and Severian period can also be divided. The sherd manufactured in the Pacatus workshop in Aquincum (sample 90) shows differences from the western sigillatae in almost every concentration defined. The differences are so significant that after the analysis of some more pieces the analytical definition of this workshop wfll not cause any difficulties. By the definition of the separational criteria the wrong attribution occurring earlier — which regarded the imported sherds of poor quality as local imitation - was revised. Summing up it can be stated that the majority of sigillata sherds found in Pannónia can - based on their chemical properties — be separated and using the analytical data we can assign to workshops even those sherds whose origin was not adequately defined by classical archaeology.