Achaeometrical Research in Hungary II., 1988

ANALYSES - POTTERY - Erzsébet JEREM - Márta BALLA - László BALÁZS: Early celtic stamped pottery in the eastern Alpine area: workshop activity and trade

3. Experimental Batches of samples were irradiated in the pool-type reactor of the Institute of Nuclear Techniques, Technical University of Budapest, with a thermal neutron flux of 2.4 x 10

12 ncm'V

1 for 8 hours. Each sample was measured twice, first after a 5-6 days cooling time and then after 25-30 days. Measurements were performed by a Canberra HPGe Well-type detector (resolution 1.95 keV, rel. efficiency 20.5%) connected to a Canberra S

100 Multichannel Analyser. For the evaluation of spectra, SAMPO 90 software was used. Standardisation was made by the single comparator method (DeCORTE, 1987), using gold as the comparator element. The thermal/epithermal flux-ratio was monitored by zirconium foils. The precision of measurements was verified by irradiating samples of the NBS SRM 1633a Coal Fly Ash standard reference material. Data on elemental concentration were processed by using multivariate statistical methods. Trace element concentrations, as vector variables, were considered to be lognormally distributed within a given group of samples (BLEBER et al., 1976). Test of normality was not performed because of the relatively small number of samples. Firstly, possible correlations between the elements measured were studied and have been found significant;in some cases. In order to remove bias caused by elemental correlations and to reduce the number of variables as well, a principal component analysis was applied. A cluster analysis was performed as well, using Euclidean distances between principal components. 4. Discussion On the basis of our analyses and calculations the previously posed questions may be answered as follows: As is shown in Fig 4, ware associated with the Sopron workshop form a major, contiguous group. Finer distinctions, however, may be made within this group, since samples from the pottery fired in Kiln No. 199 and the rest of the sherds from Sopron are somewhat separated from each other. This may be explained by the phenomenon that the rest of the samples are dispersed along a broader time interval, within which only a limited horizon corresponds to the actual use of the kiln under discussion here (since it must have been operated during a well defined period). It may be stated with certainty that the great variety of stamp decorated vessels were locally produced since they are perfectly integrated with the rest of the samples and also display similarities with each other. No significant differences have been found between ceramic vessels that were used as containers of food and beverage within the grave furniture and pottery used as cooking vessels and table services at the settlement. This supports the observation, that during the Late Iron Age, in addition to basic attire, mundane objects were also enterred with the deceased for use in the other life. Some of the comparative sample from the closer or broader environment of Sopron also confirm the results obtained by archaeological and statistical evaluations and display corresponding similarities in chemical compositions. In these cases, one may with great probability assume that they were produced in the workshop that operated in Sopron as seems to be the case with stamp-decorated vessels ornamented with "S"-shaped imprints. Earlier hypotheses concerning such similarities (JEREM 1984, Fig. 12) were convincingly supported by the perfect correspondence between a new find from Sopron (Fig.7 shoulder fragment from a bottle, also analysed by NAA) which corresponds perfectly to a vessel 89