A new model is proposed for analysis of the source clays used to create ceramics, based on geographic, petrographic, mineralogical, mineral-chemistry, and geochemical criteria. The development of this model became feasible after the discovery of a Pliocene volcanic clay horizon on NWAegina Island, Greece. The volcanic clay contains smectite, mixed-layer chlorite-smectite, biotite, and palygorskite and has greater feldspar content than the underlying Pliocene marls, which contain R0 mixed-layer illitesmectite, mica, dolomite, serpentine, talc and gypsum, and, in some places, palygorskite. The two units have distinct geochemical characteristics. In general the Pliocene volcanic clay is richer in SiO2, Al3O3, and Fe2O3 and poorer in Na2O, MgO, and P2O2 than the Pliocene marls. The Nb, Zr, Hf, Th, and rare earth element (REE) contents are also significantly greater in the Pliocene volcanic clay and comparable to those of the dacitic rocks of the island, reflecting the volcanic origin of the clay.

The proposed model was used to identify the source-clay materials that were used for the production of ceramics on the island of Aegina (Aeginetan Ware). All five criteria should be considered in any provenance study. The use of individual criteria on their own can lead to ambiguous conclusions. In the present study the geochemical criterion was particularly helpful. It provided robust evidence for the nature of the source clay. The Pliocene volcanic clay horizon and the underlying Pliocene marls are the candidate raw materials for Aeginetan Ware. Although the Pliocene marls have been invoked as raw materials for Greek Bronze Age (~3000–1100 BC) Aeginetan ceramics and are used as raw materials by modern Aeginetan ceramists, the geochemical characteristics of a large set of Bronze Age Greek Aeginetan sherds with fine and coarse fabrics coincide with those of the Pliocene volcanic clay. This comparative and cumulative evidence suggests that the Pliocene volcanic clay was the main source clay for ancient Aeginetan ceramics, regardless of the fabric (coarse or fine) and that admixture of different sources might not be necessary for fine-grained ceramics.

Thick successions of turbidites are widespread in the Ross–Delamerian and Lachlan orogens and are now dispersed through Australia, Antarctica and New Zealand. U-Pb detrital zircon age patterns for latest Precambrian, Cambrian and Ordovician metagreywackes show a closely related provenance. The latest Neoproterozoic–early Palaeozoic sedimentary rocks have major components, at c. 525, 550, and 595 Ma, i.e. about 40–80 million years older than deposition. Zircons in these components increase from the Neoproterozoic to Ordovician. Late Mesoproterozoic age components, 1030 and 1070 Ma, probably originate from igneous/metamorphic rocks in the Gondwanaland hinterland whose exact locations are unknown. Although small, the youngest zircon age components are coincident with estimated depositional ages suggesting that they reflect contemporaneous and minor, volcanic sources. Overall, the detrital zircon provenance patterns reflect the development of plutonic/metamorphic complexes of the Ross–Delamerian Orogen in the Transantarctic Mountains and southern Australia that, upon exhumation, supplied sediment to regional scale basin(s) at the Gondwana margin. Tasmanian detrital zircon age patterns differ from those seen in intra-Ross Orogen sandstones of northern Victoria Land and from the oldest metasediments in the Transantarctic Mountains. A comparison with rocks from the latter supports an allochthonous western Tasmania model and amalgamation with Australia in late Cambrian time.

We present data on the chemical and lead isotope composition of copper and bronze objects from Nuragic Sardinia. The sample suite comprises, inter alia, objects from the hoard finds at Arzachena (21 objects), Bonnanaro (10), Ittireddu (34), and Pattada (20), all in northern Sardinia. With one exception, all ingot fragments (49) consist of unalloyed copper; the exception comes from Ittireddu and contains 11 per cent tin. In contradistinction, all implements (21) are made from standard bronze with a mean tin content of 10.8 per cent. A dozen sword fragments from the Arzachena hoard, all of fairly uniform small size, are pieces of a large number of different swords. The low tin content of only about 1 per cent would have made for poor weapons, confirming the archaeological identification of the fragments as pieces of votive swords. Scrap metal from Arzachena is remarkable for its wide range of trace element contents and lead isotope abundance ratios. It is dissimilar to all other metal samples investigated, possibly representing metal from local smelting experiments using a variety of different copper ores. Lead isotope data and trace element patterns, alone or in conjunction, do not allow us to tell oxhide ingots from plano-convex (bun) ingots. Most ingot fragments have a lead isotope signature similar to those of Cypriot copper ores but there are also a number of ingots whose lead isotope fingerprints are fully compatible with them being local products. Of the bronzes, none has lead with an isotopic composition characteristic of copper ingots from Cyprus. All contain local lead, suggesting the bronze implements were manufactured locally. Isotopically-fitting lead is found in copper and lead ore deposits from the Iglesiente-Sulcis district in south-west Sardinia and from Funtana Raminosa in central Sardinia.