The end of the Late Bronze Age and the beginning of the Early Iron Age in Troia is marked by a sequence of destruction and reconstruction horizons. In the settlement phase Troia VIIb2 (ca. 1100-1000 B.C.) new vessel forms of hand-made pottery with incised and knobbed decoration appear. The same vessel forms with similar decoration can be found in several archaeological localities in Thrace and the Balkans. The primary question in investigating the Knobbed Ware of Troia and the Balkans is whether there is a possible common origin of the pottery from different regions. Was the Knobbed Ware in Troia transported from the Balkans, or do we have to do with transfer of technological solutions and the vessels were made on the spot from local material?

To answer this question several pottery and geological samples have been colleted from Troia and from archaeological sites in Bulgaria (Sborianovo, Diadovo, Ovcarovo, Chal) and Turkish Thrace (Menekşe Çatagi, etc.).

Petrographical investigation (optical microscopy incl. modal measurements) and geochemical analyses (XRF, Radiogene Isotope Analysis) have been carried out in order to group the archaeological and geological material. The volume of temper has been measured in all thin sections using a square grid. On the basis of petrographical (modal) measurements the sherds have been grouped into eight groups und several subgroups. The sherds contain mostly following minerals: quartz, primary calcite, feldspar, mica, epidote, amphibole, garnet and opaque clasts as well as rock fragments: polychrystalline quartz, magmatite, volcanite, volcanic glass, ARF, phyllite and sandstone. However, the geographical proximity and similar geology of the sites in Bulgaria where Knobbed Ware was found did not allow for differentiating clearly the petrographical groups. Further X-ray fluorescence analysis has been carried out in order to clear the grouping. Based on these data most potteries have good overlap with chemical patterns of the local sediments, so they seem to be locally produced. This observation is in accordance with archaeological theories concerning distribution of this type of pottery. It was also possible to establish that several sherds from vessels found in South-Bulgarian sites were in fact produced in North Bulgaria.

The sherds which could not be definitely grouped with the above mentioned methods have been analysed with radiogene isotopes. Data have been compared to radiogene isotope values of sediments in Bulgaria, Thrace and Troia, and locally produced ceramics in Troia. One of greatest problems posed in this work was to filter the chemical influence of temper on chemical and isotope geochemical composition. We could show that sherds containing added temper (minerals and rock fragments, e.g. volcanites, magmatites) rich in trace elements, can heavily influence the chemical composition of the raw material (clay) of the sherds, and thus complicate the interpretation of the data. Consequently, in case of coarse wares, only the combined use of petrographical and geochemical analyses can give effective provenance.