Mesolithic occupations and environments on the island of Ikaria, Aegean, Greece

The most important Mesolithic site on the Island of Ikaria, Kerame 1, extends 80 m along the sloping edge of the cliff and is up to 40 m wide. The site is a sum of repeated sojourns of Mesolithic groups that had left behind concentrations of lithic artefacts, which were subsequently displaced by post-depositional agents, fi rst of all by erosion. As a result, the site reveals now a large concentration of fi nds in Trenches E, C, and G. Moreover, post-depositional agents caused the destruction of permanent features such as the hearths associated with the various khsemenitsas, or - possibly - stone rings surrounding the dwelling structures. Only in trenches D, B and E the remains of a circular stone rings, probably around hearths, were registered. The lithic industry of Kerame 1 displays considerable similarity to the site of Maroulas on Kythnos; the techno-morphological differences are, probably, the effect of differing raw materials structure at Kerame 1 and at Maroulas. At Kerame 1, the distant interregional contacts and the infl ux of extralocal raw materials (documented by the fl ow of obsidian nodules from Melos and Yali) caused that production in a full cycle was carried out on-site. Thus, there was no specialization of lithic production, and unworked nodules of raw material were exploited in the particular social clusters in a full cycle, whose outcome were tools to be used by a given unit. Regretfully, because organic materials (also bones) have not been preserved we have no data to determine seasonality at Kerame 1. Nevertheless, we can say with all certainty that Mesolithic groups visiting Kerame 1 were mobile, which is evidenced by the network of interregional contacts. The most noticeable similarity between Kerame 1 and Maroulas can be accounted for by the chronological closeness of the two sites. The AMS determinations from Maroulas concentrate in the fi rst half of the 9th millenium cal. BC (Facorellis et al. 2010). Similarly, the dates from obsidian dehydration from Kerame 1 (if their broad standard deviation is overlooked) correspond to the fi rst half of the 9th millenium cal. BC