A statistical procedure for grouping pottery in provenance studies by chemical data is presented, which now is routinely in use in our laboratory. It is based on the Mahalanobis filter method and X2 ‐statistics, and can be used for both establishing groups and assigning single sherds to already known groups, thus replacing principal components analysis or cluster analysis and avoiding their problems in grouping pottery. The new method is able to consider correlations, uncertainties of measurement and constant shifts of the data in case of dilution. In particular, considering dilution effects results in both a better assignment to and separation of reference groups and is also equivalent to the compositional data approach, if log‐transformed data are used. Other distortions of data (e.g., mixing of clays) can also be considered.

From 38 sherds with 'corroded' surfaces, two samples per sherd were examined using Neutron Activation Analysis, one from the surface and one from the core of each sherd, in order to analyse post-depositional alterations of minor and trace elements. The most striking effect was that a leaching of Ca could be found at the surface. Another group of elements that is severely affected are the alkali metals, with Cs and Rb showing the strongest changes. A third conspicuous group of elements are the Rare Earth Elements. Finally, the effect of such alterations on a statistical data evaluation to classify pottery according to provenance is investigated.

The approaches of comparative studies and profile measurements, often used in order to detect post-depositional alterations of ceramics, have been applied simultaneously to two sets of Roman pottery, both including altered individuals. As analytical techniques Neutron Activation and X-Ray Diffraction have been used. Both approaches lead to substantially different results. This shows that they detect different levels of alteration and should complement each other rather than being used exclusively. For the special process of a glassy phase decomposition followed by a crystallization of the Na-zeolite analcime, the results suggest that it changes high-fired calcareous pottery fast and so fundamentally, that the results of various archaeometric techniques can be severely disturbed.

A new similarity measure is proposed, the ‘dilution factor spread’ or, derived from it, a ‘goodness of fit’ parameter. This has the advantage that raw data without any transformation can be used, diluted samples are recognized and errors of individual concentration values can easily be included. The use of this similarity coefficient in multivariate cluster analyses to construct dendrograms is shown and compared with the use of the well‐known similarity measure of Euclidian distance in attribute space and of the cos θ measure.

Neutron activation analysis of pottery was established at Bonn in 1983 and has since become one of the primary archaeometry-based analytical techniques at the facility. A brief history of the laboratory and a discussion of the best relative fit procedure for pottery is provided. When comparing concentration data for pottery, a best relative fit should always be considered. This mathematical procedure generally results in 'sharper' concentration patterns and improves the separability of chemically not very different compositional groups. This is demonstrated for a set of 30 Late Cypriot (Myc. IIIC1) pottery samples from Sinda, Cyprus, which allow formation of a good reference pattern for this site. Applying factors in the range of 0.82-1.43, a number of samples from Egypt and Palestine can be assigned with high probability to a Cypriot origin.