Archeometric characterization of prehistoric grindstones from Milazzo Bronze Age settlement (Sicily, Italy) Marcella Di Bella, Francesco Italiano, Maria Clara Martinelli, Paolo Mazzoleni, Simona Quartieri, Gabriella Tigano, et al. Archaeological and Anthropological Sciences ISSN 1866-9557 Archaeol Anthropol Sci DOI 10.1007/s12520-017-0483-8 1 23 Your article is protected by copyright and all rights are held exclusively by SpringerVerlag Berlin Heidelberg. This e-offprint is for personal use only and shall not be selfarchived in electronic repositories. If you wish to self-archive your article, please use the accepted manuscript version for posting on your own website. You may further deposit the accepted manuscript version in any repository, provided it is only made publicly available 12 months after official publication or later and provided acknowledgement is given to the original source of publication and a link is inserted to the published article on Springer's website. The link must be accompanied by the following text: "The final publication is available at link.springer.com”. 1 23 Author's personal copy Archaeol Anthropol Sci DOI 10.1007/s12520-017-0483-8 ORIGINAL PAPER Archeometric characterization of prehistoric grindstones from Milazzo Bronze Age settlement (Sicily, Italy) Marcella Di Bella 1,2 & Francesco Italiano 2 & Maria Clara Martinelli 3 & Paolo Mazzoleni 4 & Simona Quartieri 1 & Gabriella Tigano 5 & Alessandro Tripodo 1 & Giuseppe Sabatino 1 Received: 11 October 2016 / Accepted: 17 February 2017 # Springer-Verlag Berlin Heidelberg 2017 Abstract The results of a petrographic and geochemical study carried out on archeological grindstones allow to provide new constraints on protohistoric commercial exchanges over the Mediterranean area. Eleven grindstones, discovered in an archeological site located in Milazzo (Messina, Sicily) and dated from the Early Bronze Age, have been investigated by geochemical and petrographic techniques. The raw materials are mainly volcanic rocks characterized by calc-alkaline and K-alkaline affinities with volcanic arc geochemical signature. Only one sample, made of basalt belonging to the Naalkaline series, shows an intraplate signature. The comparison with the available literature data for similar rocks allowed constraining the volcanic origin of the exploited lavas. While the intraplate-type raw material came from Mt. Etna Volcano (Sicily), the arc-type volcanic rocks are mostly trachyandesites, basaltic andesites, and one rhyolite. Although most of them come from the Aeolian Arc, a provenance of some samples from the Aegean Arc cannot be * Marcella Di Bella mdibella@unime.it 1 Dipartimento di Scienze Matematiche e Informatiche, Scienze Fisiche e Scienze della Terra, Università degli Studi di Messina, Viale G. d’Alcontres 31, 98166 Sant’Agata, Messina, Italy 2 Istituto Nazionale di Geofisica e Vulcanologia, Sez. Palermo, Via Ugo La Malfa, 153, 90146 Palermo, Italy 3 Museo Archeologico Regionale Eoliano BLuigi Bernabò Brea^, Lipari, Aeolian Islands, Messina, Italy 4 Dipartimento di Scienze, Biologiche, Geologiche e Ambientali, Corso Italia, 57, 95129 Catania, Italy 5 Unità Operativa Beni Archeologici, Soprintendenza Beni Culturali ed Ambientali, Messina, Italy excluded. This last region could represent the most probable provenance area for the rhyolite sample. Keywords Archeometry . Early Bronze Age . Grinding tools . Volcanic rock . SEM-EDS . XRF Introduction This work aims to broaden our knowledge on archeological grindstones, made of volcanic rocks, recovered in the Bronze Age archeological site located in Piazza XXVAprile, Milazzo (Messina Province, Sicily, Italy, Fig. 1), during the excavations carried out by the Superintendence of Cultural Heritage of Messina in 2004–2005 (Tigano 2009, 2011). Grindstones are common archeological finds at almost every prehistoric site. The presence and recognition of these artifacts has a long tradition within archeology (e.g., Holmberg 1998). Even if different kinds of sedimentary, metamorphic, and basalt stones were usually reported as used raw materials, most of these artifacts were obtained from vesicular volcanic rocks, selected for their textural and physical characteristics, as high strength, hardness, and good porosity, workability, and transportability (Di Bella et al. 2016). The grinding tools appear in many shapes and sizes, representing vital objects in the daily life of the prehistoric people. In the Mediterranean area, grindstones represent a class of objects widely diffused from a geographical point of view and characterized by a relatively few morphological changes through time. The studied objects are manual grinding tools classified as of saddle quern type (Storck and Teague 1952). Numerous techno-morphological studies, often based on ethnographic data, have identified different types of the so-called saddle querns, representing the basic shapes used in the protohistoric age. Saddle querns are composed of two stones: an Author's personal copy Archaeol Anthropol Sci Fig. 1 Sketch map of the location of BPiazza XXV Aprile^ archeological site (red star) (Milazzo, NE Sicily) in the mainframe of the Aeolian Arc and Northern Sicily active tool, usually called handstone, working with a linear movement of the hands on a passive surface, formed by a grinding slab or grindstone (Curwen 1937; Storck and Teague 1952; Roux 1986; De Beaune 1989; Antonelli et al. 2004). The archeometric studies of prehistoric grinding tools from archeological sites in the Mediterranean area are important to identify the production sites and trace the historical trade routes (Childe 1943; Williams-Thorpe and Thorpe 1990, 1993; Curtis 2001; Santi et al. 2004; Williams and Peacock 2011). Williams-Thorpe and Thorpe (1993) provided a distribution map of the grinding tools in the Eastern Mediterranean Basin clearly pointing out the large amount of sites existing in Sicily with respect of the rest of Italy. Moreover, the distribution map shows that, from the Neolithic to the Roman age, grinding tools and millstones were widely marketed toward and from various provinces of the area. Many authors recognized numerous Mediterranean areas exploited for volcanic rock quarrying suitable for the production of cereal millstones especially during the Roman period. The main examples are located in Spain, Germany, France, Italy, Portugal, Morocco, Tunisia, Libya, and Turkey (Peacock 1980; Williams-Thorpe 1988; Antonelli and Lazzarini 2010, and references therein). Regarding Italy, the most important areas, from the north toward the south, were the Euganean Hills (Veneto); Vulsini Volcanic District (Latium and Umbria), especially lavas from the Orvieto region (Umbria); Somma-Vesuvius (Campania); Monte Vulture Volcano (Basilicata); Oligocene-Miocene and PliocenePleistocene Sardinian volcanic; and many areas of Sicily including Mt. Etna, Hyblean Hills, Pantelleria, Ustica, and Aeolian Islands (Ferla et al. 1984; Daniele 1997; Antonelli et al. 2000, 2001, 2004, 2005, 2014; Antonelli and Lazzarini 2010, 2012; Buffone et al. 2003; Santi et al. 2004, 2013, 2015; Gluhak and Schwall 2015). The prehistoric typologies of grinding tools have been less studied than those in the Roman period, and only a few archeometric investigations of prehistoric samples are reported in literature (Cattani et al. 1997; Antonelli et al. 2000, 2004; Author's personal copy Archaeol Anthropol Sci Lorenzoni et al. 2000). The provenance attributed by these authors to the used raw materials from the Euganean Hills, the Mt. Etna, and the Mt. Vulture has shown that in the protohistoric age, there was a well-organized trade of grindstones in the Italian Peninsula. In this context, the aim of the present work is to increase our knowledge about the provenance of grinding stones, using an approach based on mineralogical, petrographic, and geochemical investigations of the rock fragments. Main Bronze Age facies in Sicily and the BPiazza XXV Aprile^ archeological site The Early Bronze Age in Sicily is featured in numerous cultural facies spread over different areas of the Sicilian territory. The chronology of those facies is still debated especially when compared to the Italian Peninsula prehistory. The Sicilian facies are placed in the Early and Middle Bronze ages, from the end of III millennium BC to the half of II millennium BC (Bernabò Brea 1960; Tusa 1992). The fundamental features of the Early and Middle Bronze ages are the intensification of cultural and commercial exchanges among distant countries, especially the Aegean-Anatolian areas, through the strait of Messina (Marazzi 1997; Marazzi 2014). The most important facies of the Early Bronze Age in North Eastern Sicily are BCapo Graziano^ (it takes the name by the Filicudi Island, Aeolian Archipelago) and Bthe Castelluccio^ Table 1 (name taken by a settlement near Noto, Siracusa). The Capo Graziano facies testified that the Aeolian Islands experienced a new civilization during the Early Bronze Age, thanks to their favorable geographic position between the Western and the Eastern Mediterranean (Aegean-Anatolian area). This facies lasted for a long time, from 2300 BC to 1500 BC (Martinelli 2017). Besides the Filicudi (Capo Graziano) settlement, many others have been recognized in the Aeolian Islands, the most important of which are located in Stromboli, Salina, and Lipari. At the end of the Bronze Age, between 1700 BC and 1500 BC, the villages moved toward elevated and defended positions, as testified by the sites of Montagnola of Capo Graziano in Filicudi and Acropoli in Lipari (Bernabò Brea and Cavalier 1980, 1991). The Castelluccio facies, diffused in the Western and Central Sicily, is distinctively characterized by a pottery production painted in black on a red background (Cultraro 1996). Along with the painted pottery production, we found an undecorated gray pottery, mainly used for cups with handles fitted with developed appendices. Luigi Bernabò Brea (1967, 1985) recognized the gray pottery as an expression of different facies, called RodìTindari, extended across the Central and Eastern Sicily (Bernabò Brea 1967, 1985; Cavalier 1970). In the Messina strait area and in the Tyrrhenian Calabria, the archeologists have recognized a new facies, similar to Rodì-Tindari but with different types of pottery. It has been recently called facies of Messina-Ricadi or Bfacies dello Stretto^ (Procelli 2004; Martinelli et al. 2012). Information about the archeological excavation and typology of the studied grindstones Sample Document excavation Stratigraphic position Description Typology Provenance MIL1 Fragment of mortar. 23 × 20 × 12 cm Fragment of pumice? with traces of use. 14 × 11 cm Handstone Mt. Etna Handstone ? Handstone Aeolian Arc Handstone Aeolian Arc Handstone ? MIL8 In connection with US186: bottom layer with irregular stone walls and groupings of stones 15-L46 US217/II Covered by US186 that is covered by US175: gravel and land of orange-yellow color which contains US218 (circular pit stone/probable silos) 8-M46 US194 In connection with US186: bottom layer with irregular stone walls and groupings of stones 9-M46 US194 In connection with US186: bottom layer with irregular stone walls and groupings of stones 13-IL47 US194 In connection with US186: bottom layer with irregular stone walls and groupings of stones 16-pozzetto 6 US201 Inside well pottery 17-L46 US194 In connection with US186: bottom layer with irregular stone walls and groupings of stones 4-N48 US168 In connection with US124: groupings of stones MIL9 12-US168 MIL2 MIL3 MIL4 MIL5 MIL6 MIL7 1-L46 US194 MIL10 3-L45 US175 MIL11 14-N44 US124-II Fragment of grinding stone. 14 × 18 × 8 cm Fragment of grinding stone. 19 × 10 × 7 Fragment of grinding stone. 28 × 20 × 8 cm Intact. Conical shape. 16 × 13 cm The stone has no traces of use. 32 × 26 × 13 cm Fragment of grinding stone. 13.5 × 5 × 11.5 In connection with US124: groupings of stones Fragment of grinding stone with traces of use on both surface. 18 × 17 × 8 cm In connection with US124 cut 2: irregular Fragment of grinding stone. cobbled floor 15 × 9 × 4 cm Upper layer removed with 4 cuts The stone has not traces of use. 21 × 10 cm Handstone Aeolian Arc Undetermined Aeolian Arc Handstone Handstone Aegean Arc Handstone Undetermined Aeolian Arc Author's personal copy Archaeol Anthropol Sci Fig. 2 Photographs of the typologies representative of the grindstones studied here. a MIL1. b MIL5. c MIL3. d MIL9 The Bronze Age site of BPiazza XXV Aprile^ located in Milazzo (NE Sicily, Fig. 1) was an agricultural working area. The settlement was located in a saltwater environment close to the seaside, on a geological stratigraphic sequence composed of alluvial sediments (fine yellow gravel lenses and mica-rich Fig. 3 Photomicrographs (crossed nicols) of some selected grindstones samples. a MIL1: plagioclase (Pl) and clinopyroxene (Cpx) phenocrysts in a microcrystalline groundmass (gdm) mainly composed of plagioclase ± pyroxene ± olivine ± opaques. b MIL6: plagioclase (Pl) phenocrysts in a microcrystalline intersertal groundmass. c MIL4: clinopyroxene ± plagioclase and olivine phenocrysts in a vitrophiric groundmass. d MIL9: plagioclase ± brown amphibole microphenocrysts in a cryptocrystalline groundmass gray powdery sands) coming from the estuary of the Mela River. The settlement age, on the basis of archeological remains and of the pottery typologies, has been attributed to the Messina-Ricadi facies, dated to the end of the Early Bronze Age, in the half of II millennium BC (Martinelli 2009; Author's personal copy Archaeol Anthropol Sci Martinelli and Tigano 2012; Tigano 2011). Over the Milazzo BPiazza XXV Aprile^ archeological site, a large amount of grinding stones were recovered as discarded materials or recycled as building stones besides other remains of structures such as cobbled floors, storage pits, and little wells made of two or three stacked jars used to collect the groundwater. following Franzini and Leoni (1972). The analytical precision is ±0.3% for Si and ±0.1% for other major elements. For the trace elements, the analytical precision is ±5% for Rb and Sr and ±10% for the others. Results Sampling and analytical methods Petrography From the lithic material composed of 11 fragments of grinding stones, one piece of pumice, one fragment of mortar, and a group of raw obsidians, a sample suite of 11 fragments (labeled from MIL1 to MIL11; Table 1) was selected from the grindstones (see Fig. 2) upon the permission of the Messina Superintendence of Cultural Heritage. Despite the evident volcanic origin of the sample suite, they underwent laboratory analyses for a correct volcanological and petrographic classification. A slice was cut from every fragment to produce a thin section for the optical observations. The rest of the sample was cleaned in an ultrasonic bath, dried, crushed in a jaw crusher, and powdered in an agate mill, except MIL8 and MIL10 due to their very low amount. The thin sections underwent petrographic and mineralogical analyses by optical and environmental scanning electron microscopy. The powders have been used for major and trace element determination by XRF analysis. Most of the analytical investigations were performed at the CERISI laboratories and the MIFT Department (Scienze Matematiche e Informatiche, Fisiche e Scienze della Terra) of the Messina University. SEM-EDX analyses were carried out by an ESEM-FEI Inspect-S electron microscope coupled with an Oxford INCA PentaFETx3 EDX spectrometer and a Si(Li) detector equipped by an ultra thin window ATW2, using a resolution of 137 eV at 5.9 keV (Mn Kα1). The spectral data were acquired in environmental scanning electron microscope (ESEM) conditions at a working distance of 10 mm with an acceleration voltage of 20 kV and counting time of 60 s, at approximately 3000 cps with dead time below 30%. The results were processed by INCA Energy software use in the XPP matrix correction scheme (Pouchou and Pichoir 1984, 1985). The whole rock chemical XRF analyses (major and trace elements) were carried out on powder pellets at the Modena and Reggio Emilia University, using a wavelength-dispersive automated Philips PW1400 spectrometer. Magnesium and sodium concentrations were determined by atomic absorption spectrophotometry and flame emission on a sample solution obtained after perchloric and hydrofluoric acid attack. Iron content was determined by titration after rapid HF-H2SO4 attack. Loss on ignition (LOI) was determined by the weight loss after heating at 950 °C. The concentrations of each element were calculated upon the correction of the matrix effects All the collected samples (Table 1) are medium to high porphyritic lavas, highly vesiculated and characterized by fresh appearance and dark gray color (Figs. 2 and 3). They show several generations of phenocrysts mainly involving plagioclase, pyroxene, and minor olivine. The groundmass, characterized by microcrystalline intersertal or cryptocrystalline and in some cases vitrophiric structure, consists of the same phases present as phenocrysts with prevalence of plagioclase microliths frequently showing trachitic texture. The composition of Fig. 4 a, b Classification diagrams of feldspars (Deer et al. 1963) and pyroxenes (Morimoto et al. 1988) of the studied samples Author's personal copy Archaeol Anthropol Sci Table 2 Representative compositions of plagioclase, pyroxene, and olivine phenocrysts present in the studied samples Sample lithology Mineral MIL1 MIL2 MIL3 MIL7 Hawaiite Basaltic andesite Trachyandesite MIL9 Rhyolite SiO2 Al2O3 Plagioclase 54.38 29.80 51.96 31.71 49.19 31.26 48.55 32 57.41 26.06 56.93 27.03 56.09 27.34 57.63 25.97 50.92 30.43 50.87 31.64 FeO CaO Na2O K2O 0.52 10.61 4.40 0.30 0.47 12.77 3.03 0.06 1.28 16.08 1.88 0.3 1.34 16.35 1.49 0.28 0.45 8.71 6.12 1.25 0.39 9.26 5.52 0.87 0.64 10.39 4.85 0.69 0.68 10.03 4.8 0.9 0.90 14.78 2.62 0.34 0.78 14.14 2.36 0.22 TOTAL Cations 100.01 100.00 99.99 100.01 100 100 100 100.01 99.99 100.01 Si 2.4362 2.3377 2.2347 2.2046 2.5790 2.5522 2.5153 2.5778 2.3082 2.2965 Al Fe 3+ Ca Na 1.5734 0.0390 1.6814 0.0354 1.6737 0.0973 1.7126 0.1018 1.3798 0.0338 1.4282 0.0292 1.4450 0.0480 1.3691 0.0509 1.6257 0.0682 1.6834 0.0589 0.5093 0.3822 0.6156 0.2643 0.7827 0.1656 0.7955 0.1312 0.4192 0.5330 0.4448 0.4798 0.0000 0.4992 0.0000 0.4807 0.7178 0.2303 0.6839 0.2066 K 0.0171 0.0034 0.0173 0.0162 0.0716 0.0498 0.4217 0.4163 0.0197 0.0127 Z X End members 4.04863 0.90861 4.05447 0.88332 4.0057 0.9657 4.0190 0.9429 4.0486 0.9086 4.0932 0.8355 4.0083 0.9604 3.9977 0.9483 4.0021 0.9677 4.0388 0.9032 Ab An Or 47.41 36.05 16.54 29.92 69.69 0.39 17.15 81.05 1.80 13.91 84.37 1.72 42.06 56.05 1.89 41.75 56.68 1.57 43.91 51.98 4.11 43.90 50.69 5.42 23.79 74.17 2.03 22.87 75.73 1.40 Mineral Na2O MgO Al2O3 SiO2 K2O CaO TiO2 MnO Pyroxene 0.67 11.75 3.83 50.71 0.21 21.50 1.17 0.32 0.55 12.75 4.69 51.80 0.32 21.55 0.88 0.21 0.87 13.59 5.19 54.36 0.29 15.67 0.42 0.30 0.85 13.71 3.55 64.36 0.40 4.20 0.24 0.41 0.63 14.76 3.83 53.11 0.28 19.88 0.67 0.15 0.70 22.10 2.67 56.16 0.45 1.96 0.30 0.89 0.60 10.82 2.10 60.47 0.32 14.90 0.51 0.26 0.52 10.69 2.22 54.49 0.23 18.15 0.62 0.38 0.52 13.86 2.84 53.42 0.16 19.58 0.53 0.51 0.57 13.22 2.75 53.93 0.00 20.12 0.46 0.00 FeO Total 9.85 100.01 7.27 100.02 9.32 100.01 12.28 100.00 6.69 100.00 14.76 99.99 10.02 100.00 12.70 100.00 8.59 100.01 8.96 100.01 1.9041 0.0330 0.1695 0.3093 0.0102 0.6577 0.8649 0.0488 0.0101 4.0076 1.9149 0.0245 0.2043 0.2248 0.0066 0.7026 0.8535 0.0394 0.0151 3.9857 1.9809 0.0115 0.2229 0.2840 0.0093 0.7383 0.6118 0.0615 0.0135 3.9336 2.2452 0.0063 0.1460 0.3583 0.0121 0.7130 0.1570 0.0575 0.0178 3.7131 1.9453 0.0185 0.1653 0.2049 0.0047 0.8059 0.7801 0.0447 0.0131 3.9825 2.0226 0.0081 0.1133 0.4445 0.0271 1.1865 0.0756 0.0489 0.0207 3.9474 2.1739 0.0138 0.0890 0.3012 0.0079 0.5799 0.5739 0.0418 0.0147 3.7961 2.0330 0.0174 0.0976 0.3963 0.0120 0.5946 0.7255 0.0376 0.0109 3.9250 1.9722 0.0147 0.1236 0.2652 0.0159 0.7628 0.7745 0.0372 0.0075 3.9737 1.9885 0.0128 0.1195 0.2763 0.0000 0.7267 0.7948 0.0407 0.0000 3.9593 47.41 36.05 16.54 47.93 39.45 12.62 37.44 45.18 17.38 12.78 58.05 29.17 43.56 45.00 11.44 4.43 69.52 26.05 39.44 39.85 20.70 42.27 34.64 23.09 42.97 42.32 14.71 44.21 40.42 15.37 Cations Si Ti Al Fe2+ Mn Mg Ca Na K Total End members Wo En Fs Author's personal copy Archaeol Anthropol Sci Table 2 (continued) Sample lithology MIL1 MIL2 MIL3 MIL7 Hawaiite Basaltic andesite Trachyandesite Mineral SiO2 Al2O3 Olivine 45.01 0.00 48.14 0.00 40.76 1.72 41.35 1.69 41.75 1.45 41.07 1.52 FeO MnO 13.30 0.00 13.25 0.32 22.81 0.65 22.36 0.52 22.63 0.61 23.79 0.64 MgO 41.34 37.98 32.97 33.07 32.87 32.01 CaO Total 0.17 99.82 0.35 99.72 0.79 99.70 0.72 99.71 0.77 100.11 0.59 99.59 Fe2+ Mn Mg Ca Total 1.1303 0.0000 0.2793 1.2089 0.0000 0.2782 1.0235 0.0509 0.4790 1.0384 0.0500 0.4695 1.0480 0.0430 0.4650 1.0310 0.0450 0.5000 0.0000 1.5476 0.0046 2.9617 0.0000 1.4218 0.0094 2.9183 0.0138 1.2343 0.0213 2.8263 0.0111 1.2380 0.0194 3.7131 0.0140 1.2310 0.0210 2.8310 0.0013 1.1980 0.0160 2.8030 End members Fo 84.71 83.63 71.47 72.04 71.57 70.04 Fa Tp 15.29 0.00 16.37 0.00 27.73 0.80 27.32 0.64 27.64 0.79 29.20 0.76 MIL9 Rhyolite Cations Si Al the main analyzed mineral phases (phenocrysts of plagioclases and pyroxenes) is reported in Fig. 4a, b. The plagioclase phenocrysts are from euhedral to subheuedral, geminated and with sieve texture sometimes characterized by a large corroded, often fractured appearance. Plagioclase is prevalently labradoritic to bytownitic up to andesinic (MIL3); a few sanidinic terms were found in sample MIL7 (Fig. 4a). The clinopyroxene phenocrysts are greenbrown, zoned, often twinned, in some cases resorbed with inclusions of opaques, and frequently form glomeroporphyritic aggregates with plagioclase and/or olivine. They are generally diopside or augite, but MIL2 shows a pigeonite-like composition. Hyperstene phenocrysts occur in all the analyzed samples but for MIL1 (Fig. 4b). Olivine phenocrysts are heuedralsubheuedral and characterized by normal zoning with Mg-rich cores (Fo ≈ 70–80%) and rims enriched in fayalitic component (Fo ≈ 60–50%). Small and microolivine phenocrysts, marked by low forsterite content (Fo 27–22), frequently occur in glomeroporphyritic aggregates besides plagioclase and clinopyroxene. Concerning the groundmass, samples MIL3, MIL4, MIL6, MIL7, MIL8, and MIL11 show a dark color and glassy appearance. Contrastingly, MIL1, MIL2, MIL9, and MIL10 exhibit trachitic texture, with abundant and frequently oriented feldspar microliths (6–8 mm sized). In all samples, groundmass is formed by plagioclase > Fe–Ti oxides > clinopyroxene ± olivine. Geochemistry Major oxides and trace element compositions are reported in Tables 2 and 3. Samples MIL2, MIL3, MIL4, MIL5, MIL6, MIL7, MIL8, MIL9, and MIL11 show potassium alkaline affinity while MIL1 is characterized by sodium affinity, according to Le Maitre (2002; Na2O–2 > K2O) classification. The TAS classification (Le Bas et al. 1986; Fig. 5a) highlights that, among the potassium alkaline samples, MIL3, MIL4, and MIL7 are trachyandesite; MIL2, MIL5, MIL6, and MIL11 are basaltic andesites; and MIL9 is a rhyolite. The sodic alkaline sample MIL1 falls in the hawaiite area. Moreover, on the SiO2 vs K2O diagram (Fig. 5b), MIL2, MIL5, and MIL9 fall in the field of calc-alkaline series; MIL3, MIL6, and MIL11 plot in the area of high-K calc-alkaline rocks; and MIL4 and MIL7 in the shoshonitic one. In the spider diagrams of Fig. 5c, the abundances of incompatible trace elements were normalized with respect to the primordial mantle composition (McDonough et al. 1992). The hawaiite sample MIL1 shows a bell-shaped pattern, with positive anomaly of Nb and a general enrichment in highfield-strength (HFS) elements, testifying an intraplate environment provenance of this rock (Fig. 5c; Peccerillo 2005). Conversely, samples belonging to potassic series show higher abundances of LILE and depletion in HFSE with negative anomaly of Nb typical of volcanic arc magmas (Fig. 5c; Author's personal copy Archaeol Anthropol Sci Table 3 XRF major (weight %) and trace element (ppm) data of the studied grindstone samples Samples MIL1 MIL2 MIL3 MIL4 MIL5 MIL6 MIL7 MIL9 MIL11 Major oxides SiO2 TiO2 Al2O3 Fe2O3 49.13 1.6 18.25 9.35 55.75 0.65 19.27 8.82 56.25 0.76 19.36 6.09 55.49 0.83 19.68 5.98 wt% 55.05 0.65 19.37 8.42 55.25 0.75 18.56 8.58 55.55 0.84 19.65 6 73.97 0.34 12.46 3.2 54.06 0.69 18.01 8.65 MgO Na2O K2O 5.58 3.97 1.26 2.7 2.83 1.17 2.23 4.13 3.03 2.54 3.37 3.46 3.9 2.5 1.52 3.47 2.3 1.97 2.47 3.34 3.45 1.32 2.79 1.72 4.18 2.61 1.86 CaO P2O5 8.54 0.56 7.43 0.13 5.79 0.41 7.21 0.42 7.15 0.16 7.28 0.21 7.2 0.42 3.27 0.17 7.38 0.19 MnO LOI Trace elements Ni 0.12 1.64 0.14 1.1 0.12 1.82 0.09 0.92 0.11 1.52 0.09 0.98 0.06 0.7 0.13 2.25 51 9 12 14 0.13 1.14 ppm 21 19 15 4 11 Co Cr V Ce Nd Ba La Nb 35 90 218 139 53 719 75 45 27 18 264 38 19 412 20 3 17 17 145 122 52 1597 69 24 20 28 205 106 43 1409 58 24 30 55 228 43 22 463 27 6 34 61 245 55 28 589 33 8 19 28 205 102 44 1414 58 22 7 8 83 33 16 389 20 6 30 25 215 59 27 633 32 8 Zr 240 91 350 304 127 157 118 89 153 Y Sr Rb 22 1110 23 17 668 33 26 625 71 24 573 110 18 640 40 18 759 55 23 544 107 9 357 42 18 811 52 Pb 11 10 25 25 11 17 22 6 8 As Zn Cu 0 92 42 0 80 74 6 68 39 6 61 57 0 78 71 0 77 66 6 53 56 0 32 20 6 68 51 Peccerillo 2005). The Zr vs Ti (Pearce and Cann 1973) discrimination diagram (Fig. 5d) providing constraints on the tectonic environment shows that sample MIL1 falls in the area of within-plate or anorogenic volcanic and samples MIL2, MIL5, MIL6, MIL7, MIL8, MIL9, and MIL11 fall in the area of subduction-related basalts, while MIL3 and MIL4 fall outside the two areas. Discussion The contribution of recent and former studies allowed developing a petrological database on the provenance areas of volcanic grindstones and millstones used in Mediterranean countries from the Neolithic to the Roman ages (Williams-Thorpe et al. 1991; Antonelli and Lazzarini 2010, and references therein). The probable origin of the Milazzo BPiazza XXV Aprile^ grindstones has been constrained by the geo- mineralogical signatures of the collected samples, according to the discriminative geochemical diagrams used in igneous petrology. The analytical results of the used raw materials are compared with those of similar rocks listed in the GEOROCK database (Peccerillo 2005; Francalanci et al. 2007) with the aim to define their provenance area. The MIL1 sample, classified as intraplate or anorogenic, was compared with Italian rocks characterized by the same geochemical signature. In Italy, sodic alkaline intraplate rocks are present in Sardinia, Sicily (Mt. Etna, Hyblean Plateau), islands and seamounts of the Sicily Channel and Ustica, and the old volcanic districts of the Venetian Province and of Punta delle Pietre Nere (Apulia). A first discrimination is provided by the Ba-Sr classification diagram proposed by Antonelli and Lazzarini (2010) that includes data of Mt. Etna basalts, hawaiites, and other rocks from Pantelleria, Ustica, and the Hyblean Plateau. The Ba-Sr classification diagram (Fig. 6a) is here improved with the Author's personal copy Archaeol Anthropol Sci Fig. 5 Geochemical diagrams used to classify all the studied grindstone samples. a Total alkali vs silica diagram (Le Bas et al. 1986). b SiO2 vs K2O diagram (Peccerillo and Taylor 1976) on which the compositional fields of the Aeolian Islands rocks were plotted for comparison (data from Peccerillo 2005). c Spider diagrams of incompatible trace elements normalized to the primordial mantle of McDonough et al. (1992); Zr vs Ti tectonic discrimination diagrams (Pearce and Cann 1973). Blue circle = Na hawaiite sample, red circles = other K samples addition of data of volcanic rocks from the Sicily Channel (Peccerillo 2005) and Euganean Hills (Milani et al. 1999). In Fig. 6a, the MIL1 sample falls in the area of Mt. Etna products in agreement with the petrographic and geochemical signatures of Mt. Etna volcanic rocks (Gillot et al. 1994; Corsaro and Cristofolini 2000; Branca et al. 2004; Peccerillo 2005). The Etna hawaiites, compositionally similar to the studied MIL1 sample, generally show a gray to dark-gray color; they are slightly vesicular and seriate and often characterized by a 30–50% IP. As summarized by some authors (Antonelli et al. 2005; Antonelli and Lazzarini 2010; Di Bella et al. 2016, and references therein), they are characterized by low TiO 2 (<2 wt%) and high Sr (>1000 ppm) and Ba (>700 ppm) geochemical signature. Furthermore, those alkaline rocks show a bell-shaped pattern of primordial mantle-normalized incompatible elements characterized by negative anomalies of K, Hf, and Ti and positive anomalies of Nb and Ta (Peccerillo 2005). The similarity between Mt. Etna hawaiites and MIL1 is well evidenced by LILE/HFSE, LILE/LILE, and HFSE/HFSE ratios (for Mt. Etna Ba/Zr = 2–6, Rb/Nb = 0.5–0.7, and Ba/ La = 9–12; for MIL1 Ba/Zr ~ 2–3, Rb/Nb ~ 0.51, and Ba/ La ~ 9) as shown in Fig. 6b–e. Samples MIL3, MIL4, MIL5, MIL6, MIL7, MIL9, and MIL11, marked by a subduction-related geochemical signature, were compared with data of the most probable provenance areas represented by the Western, central, and eastern sectors of the Aeolian volcanic arc (after Peccerillo 2005 and Francalanci et al. 2007). The western portion of the arc, including Alicudi, Filicudi, and Salina Islands (Fig. 1), is composed by mafic, and subordinately silicic, calc-alkaline rocks. The central portion is represented by Vulcano and Lipari Islands (Fig. 1) characterized by calc-alkaline to shoshonitic mafic to silicic rocks. The eastern Panarea and Stromboli Islands (Fig. 1) consist of calc-alkaline to potassic alkaline rocks (Peccerillo 2005; Francalanci et al. 2007). Despite the different petrographic classification, the Aeolian Islands volcanics show very similar geochemical features in terms of major and trace elements. As shown in Fig. 5b (K2O vs SiO2), MIL2 and MIL5 are medium K calc-alkaline basaltic andesites and fall in the overlapping Author's personal copy Archaeol Anthropol Sci Fig. 6 Binary diagrams used to discriminate the provenance of hawaiite sample MIL1 (blue circle). a Sr vs Ba plot after Antonelli and Lazzarini (2010), modified in this study. b–e Nb, Rb/Nb, Zr, and Rb/Ce vs MgO plots on which the composition of the hawaiite sample was compared with literature data (from Peccerillo 2005) of Mt. Etna volcanics field of the products from Salina, Lipari, Panarea, and Alicudi, whereas MIL6 and MIL11 are high-K basaltic andesites and plot in the compositional fields of rocks from Filicudi, Stromboli, Panarea, and Salina. The trachyandesite MIL3 falls in the high-K overlapping field of Panarea, Vulcano, and Lipari rocks, while MIL4 and MIL7 fall in the shoshonitic field of Vulcano, Lipari, Stromboli, and Panarea rocks. It is worth noticing that MIL9 plots outside the compositional range of the Aeolian Islands rocks and the discrimination provided by the major elements is not enough to identify the provenance from one of the islands of the Arc. With the aim to better constrain the provenance area of the raw rocks, further classification diagrams based on the Author's personal copy Archaeol Anthropol Sci Fig. 7 Binary diagrams used to discriminate the provenance of volcanic arc samples. a Ce vs Nb plot (this study) on which the compositional fields of rocks from Aegean (dotted line area) and Aeolian Arcs (gray area) were plotted for comparison (Peccerillo 2005; GEOROCK database). b Sr vs Rb plot by Williams-Thorpe and Thorpe (1993) on which the gray area of the Aeolian Islands compositional field was added in this work. c Ba/Sr and Rb/Nb vs SiO 2 plots on which the trachyandesite samples (MIL3, MIL4, and MIL7) are plotted in comparison with literature data (from Peccerillo 2005) of Vulcano, Lipari, Panarea, and Stromboli rocks Ce, Nb, Sr, and Rb trace elements have been used. Figure 7a, b displays data from either the Aeolian and Aegean arc volcanoes. On the Ce vs Nb plot (Fig. 7a), samples MIL3, MIL4, and MIL7 fall inside the area of the Aeolian Arc whereas the other samples fall over the overlapping area of the two arcs. Contrastingly, the Sr-Rb diagram of Fig. 7b (WilliamsThorpe and Thorpe 1993) discriminates samples MIL6 and MIL11 falling in the area of Aeolian Arc rocks from MIL2 and MIL5 falling in the Aeolian-Aegean overlapping area. On the Ba/Sr and Rb/Nb vs SiO2 diagrams (Fig. 7c, d) where rocks from Stromboli, Panarea, Lipari, and Vulcano are plotted, the samples MIL3, MIL4, and MIL7 (shoshonitictrachyandesites) exhibit a composition very similar to the Stromboli rocks. Sample MIL9, finally, plots over the rocks of the Aegean Arc, in particular near the Santorini field (Fig. 7b). As a matter of fact, our results highlight the reasonable possibility that the grinding tools recovered on the Sicily mainland were imported either from the Aeolian or from the Aegean regions. Concluding remarks The results of the present study highlighted that the raw rocks used to build the analyzed prehistoric grindstones mainly derive from volcanic arc areas and, subordinately, from the intraplate environment. Concerning the subduction-related samples, an importation from the Aeolian Islands seems to be very probable, simply on the basis of the geographical location of Milazzo (Fig. 1). The obtained results clearly indicate a provenance from the Aeolian Islands for MIL3, MIL4, MIL7, MIL6, and MIL11 samples. Contrastingly, sample MIL9—falling outside the compositional field of Aeolian Arc rocks—shows geochemical affinity with rocks from the Aegean Region, probably from Santorini. Finally, the results cannot completely rule out that other samples, especially MIL2 and MIL5, had been imported from the Aegean region, although their composition well fits with rocks from the Aeolian Islands (Panarea, Lipari, Salina, and Alicudi). The proposed results testify that, during the Early Bronze Age, rocks from the Aeolian Arc were exploited for grindstone Author's personal copy Archaeol Anthropol Sci manufacturing. In this age, the Aeolian Islands played a prominent role as a strategic point toward the Southern Tyrrhenian Sea. Evidences of the importance of this area during the Bronze Age are represented by the Capo Graziano facies, with the large number of new settlements widely diffused over all of the islands (the most important are located in Lipari and Filicudi). The object of the exchange with the Aegean area through a Mediterranean Sea route is not yet fully understood. It might be linked to the search for strategic raw materials, such us metals, by the emerging Mycenaean warrior élites (Jones et al. 2014). The trade organization included valuable products (pottery, jewelries) or carrying products (such as oil and ointments). The presence in Milazzo of stone materials from the Aegean area could be perhaps explained as the result of an Bindirect^ exchange between the Aeolian communities and inhabitants of the Sicilian coast. For sample MIL1, the only intraplate-related rock, we suggest Mt. Etna as an exploited source, since petrographic and geochemical features perfectly fit with those of the mafic rocks from this area (Cristofolini and Romano 1982; Cristofolini et al. 1991; Corsaro and Cristofolini 1996). Firstly, Williams-Thorpe (1988) and, after, Renzulli et al. (2002a, b) suggested that the BFratelli Pii^ quarry (from 693 AD) was the volcanic site commonly used to work hawaiitic millstones. It is certain that Sicily, since the prehistory, provided abundant raw materials to manufacture the most primitive grindstones and the more technological millstones of Roman age (Ferla et al. 1984; Williams-Thorpe 1988; Antonelli et al. 2004, 2005; Antonelli and Lazzarini 2012; Di Bella et al. 2016). Our results, based on the studied archeological findings, confirm that in the Bronze Age, the contacts between people of eastern Sicily/Aeolian Islands and the Aegean ones were very active. Acknowledgements The authors thank Simona Bigi (XRF Laboratory of the University of Modena and Reggio Emilia) for the chemical analyses of the major elements. References Antonelli F, Lazzarini L (2010) Mediterranean trade of the most widespread Roman volcanic millstones from Italy and petrochemical markers of their raw materials. J Archaeol Sci 37:2081–2092 Antonelli F, Lazzarini L (2012) The first archaeometric characterization of Roman millstones found in the Aquileia archaeological site (Udine, Italy). Archaeometry 54:1–17 Antonelli F, Nappi G, Lazzarini L (2000) Sulla «pietra da mole» della regione di Orvieto. Caratterizzazione petrografica e studio archeometrico di macine storiche e protostoriche dall’Italia centrale. Atti del I° Congresso Nazionale di Archeometria (AIAr), Pàtron Editore, pp.195–207 Antonelli F, Nappi G, Lazzarini L (2001) Roman millstones from Orvieto (Italy): petrographic and geochemical data for a new archaeometric contribution. Archaeometry 43:167–189 Antonelli F, Bernardini F, Capedri S, Lazzarini L, Montagnari Kokelj E (2004) Archaeometric study of protohistoric grinding tools of volcanic rocks found in the karst (Italy–Slovenia) and Istria (Croatia). Archaeometry 46:537–552 Antonelli F, Lazzarini L, Luni M (2005) Preliminary study on the import of lavic millstones in Tripolitania and Cyrenaica (Lybia). J Cult Herit 6:137–145 Antonelli F, Columbu S, De Vos M, Andreoli M (2014) An archaeometric contribution to the study of ancient millstones from the Mulargia area (Sardinia, Italy) through new analytical data on volcanic raw material and archaeological items from Hellenistic and Roman North Africa. J Archaeol Sci 50:243–261 Bernabò Brea L (1960) La Sicilia prima dei Greci. II edizione, Il Saggiatore, Milano Bernabò Brea L (1967) La necropoli di Longane. Bullettino Paletnologia Italiana 76:181–254 Bernabò Brea L (1985) Gli Eoli e l’inizio dell’età del bronzo nelle isole Eolie e nell’Italia meridionale. AION, Napoli Bernabò Brea L, Cavalier M (1980) L’acropoli di Lipari nella preistoria, Meligunis Lipara IV. Flaccovio, Palermo Bernabò Brea L, Cavalier M (1991) Filicudi: insediamenti dell’età del bronzo, Meligunìs Lipára VI. Regione Siciliana, Assessorato Beni Culturali, Palermo Branca S, Coltelli M, Groppelli G (2004) Geological evolution of Etna Volcano, in Calvari S., A. Bonaccorso A., Coltelli M., Del Negro C. and Falsaperla S. (Eds). Etna Volcano laboratory, Geophysical Monographs AGU 143:49–63 Buffone L, Lorenzoni S, Pallara M, Zanettin E (2003) The millstones of ancient Pompei: a petro-archaeometric study. Eur J Mineral 15:207–215 Cattani M, Lazzarini L, Falcone R (1997) Macine protostoriche dall’Emilia e dal Veneto: note archeologiche, caratterizzazione chimico-petrografica e determinazione della provenienza. Padusa, pp:105–137 Cavalier M (1970) La stazione preistorica di Tindari. Bullettino Paletnologia Italiana 79:61–94 Childe VG (1943) Rotary querns on the continent and in the Mediterranean basin. Antiquity 17:19–26 Corsaro RA, Cristofolini R (1996) Origin and differentation of recent basaltic magmas from Mount Etna. Mineral Petrol 57:1–21 Corsaro RA, Cristofolini R (2000) Subaqueous volcanism in the Etnean area: evidence for hydromagmatic activity and regional uplift inferred from the Castle Rock of Acicastello. J Volcanol Geotherm Res 95:209–225 Cristofolini R, Romano R (1982) Petrologic features of the Etnean volcanic rocks. Memorie della Società Geologica Italiana 23:99–115 Cristofolini R, Corsaro RA, Ferlito C (1991) Variazioni petrochimiche nella successione etnea. Un riesame in base a nuovi dati da campioni di superfici e da sondaggi. Acta Vulcanol 1:25–37 Cultraro M. (1996) La facies di Castelluccio, in Cocchi Genick D. (ed.) L’antica età del Bronzo in Italia, Atti del congresso, Viareggio 1995, Firenze: 163–174. Curtis RI (2001) Ancient food technology. Brill, Boston Curwen EC (1937) Querns in antiquity 11:133–151 Daniele D (1997) Studio chimico-petrografico e individuazione delle aree di provenienza del materiale lavico delle macine di Entella. Atti Seconde giornate internazionali di studi sull’area Elima. CNR, Pisa, Gibellina, 22–26 ottobre 1994. ISBN 88-7642-071-1 De Beaune SA (1989) Essaid’une classification typologique des galets et plaquettes utilisés au paléolithique. Gallia Préhistoire 31:27–64 Deer WA, Howie RA, Zussman J (1963) Rock forming mineral. Longman Group Ltd., London, England, 435 pp Di Bella M, Mazzoleni P, Russo S, Sabatino G, Tigano G, Tripodo A (2016) Archaeometric characterization of Roman volcanic millstones from Messina territory (Sicily, Italy). Period Mineral 85:69–81 Author's personal copy Archaeol Anthropol Sci Ferla P, Alaimo R, Falsone G, Spatafora F (1984) Studio petrografico delle macine di età arcaica e classica da Monte Castellazzo di Poggioreale. Sicilia Archeol 56:25–52 Francalanci L, Vougioukalakis GE, Perini G, Manetti P (2007) A westeast traverse along the magmatism of the south Aegean volcanic arc in the light of volcanological, chemical and isotope data. Developments in Volcanology 7:65–111 Franzini M, Leoni L (1972) A full matrix correction in X-ray fluorescence analysis. Att Soc Tosc Sc Nat Mem 79:7–22 The database GEOROC (Geochemistry of Rocks of the Oceans and Continents), Max Plank Institute - http://georoc.mpch-mainz.gwdg.de/georoc/ Gillot PY, Kieffer G, Romano R (1994) The evolution of Mount Etna in the light of potassium-argon dating. Acta Vulcanol 5:81–87 Gluhak TM, Schwall C (2015) Provenance analyses of the volcanic rock grinding stones from the greek colony of Selinunte, Sicily (Italy): constraints and possibilities. Archaeometry 57:246–268 Holmberg C (1998) Prehistoric grinding tools as metaphorical traces of the past Le Bas MJ, Le Maitre RW, Streckeisen A, Zanettin B (1986) A chemical classification of volcanic rocks based on the total-alkali-silica diagram. J Petrol 27:745–750 Le Maitre RW (2002) Igneous rocks: a classification and glossary of terms: recommendations of the International Union of Geological Sciences Subcommission on the Systematics of Igneous Rocks. Cambridge University Press, Cambridge, 236 pp Lorenzoni S, Pallara M, Venturo D, Zanettin E (2000) Volcanic rock Bronze Age millstones of Apulia, Southern Italia: lithology and provenance. Eur J Mineral 12:877–882 Marazzi M (1997) I contatti transmarini nella preistoria siciliana, in Tusa S (ed.) BPrima Sicilia^, Ediprint: 364–374 Marazzi M (2014) Prima dei Fenici: i Micenei nel Mediterraneo fra espansione e collasso. In: Accardi A (ed) Ricordando Braduel. Mediterraneo un m are condivis o, R egione Siciliana. Soprintendenza del Mare, Palermo, pp 65–86 Martinelli MC (2009) L’insediamento specializzato in Piazza XXVAprile a Milazzo (ME). In: Tigano G (ed) BMilay II^, Regione Siciliana, Assessorato Beni Culturali e Ambientali, Soprintendenza BCA di Messina. Servizio per i Beni Archeologici, Sicania-Messina, pp 175–186 Martinelli M.C. (2017), Updates on the cultural and chronological framework of the prehistory and protohistory of the Aeolian Islands: from the first settlement to the end of the villages, in Atti Convegno internazionale BUbi minor. le isole minori del Mediterraneo centrale dal Neolitico ai primi contatti coloniali^, Capri 26–30 ottobre 2014, Scienze dell’Antichità 2016 M a r t i n e l l i M C a n d Ti g a n o G ( 2 0 1 2 ) U n s i s t e m a d i approvvigionamento idrico nell’età del Bronzo, Atti XLI Riunione Scientifica dell’Istituto Italiano di Preistoria e Protostoria BDai Ciclopi agli Ecisti. Società e territorio nella Sicilia preistorica e protostorica^, 16–19 novembre 2006, San Cipirello (PA) CD-ROM pp. 1301–1304 Martinelli MC, Procelli E, Pacciarelli M, Cavalier M (2012) L’età del Bronzo antica e media nella Sicilia orientale e nella zona dello Stretto di Messina, in Atti XLI Riunione Scientifica dell’Istituto Italiano di Preistoria e Protostoria BDai Ciclopi agli Ecisti. Società e territorio nella Sicilia preistorica e protostorica^ San Cipirello (PA) 16–19 novembre 2006, pp. 157–186 McDonough WF, Sun SS, Ringwood AE, Jagoutz E, Hofmann AW (1992) K, Rb and Cs in the earth and moon and the evolution of the earth’s mantle. Geochim Cosmochim Acta 56:1001–1012 Milani L, Beccaluva L, Coltorti M (1999) Petrogenesis and evolution of the Euganean magmatic complex, Veneto region, NE Italy. Eur J Mineral 11:379–399 Morimoto N, Fabries J, Ferguson AK, Ginzburg IV, Ross M, Seifert FA, Zussman J, Aoki K, Gottardi D (1988) Nomenclature of pyroxenes. Am Min 62:53–62 Peacock DPS (1980) The Roman millstone trade: a petrological sketch. World Archaeol 12:43–53 Pearce JA, Cann JR (1973) Tectonic setting of basic volcanic rocks determined using trace element analyses. Earth Planet Sci Lett 19:290–300 Peccerillo A (2005) Plio-quaternary volcanism in Italy: petrology, geochemistry, geodynamics. Springer, Heidelberg, 365 pp Peccerillo A, Taylor SR (1976) Geochemistry of Eocene calc-alkaline volcanic rocks from the Kastamonu area, northern Turkey. Contrib Mineral Petrol 58:63–81 Pouchou JL, Pichoir L (1984) Possibilités d’analyseen profondeur à la microsonde électronique. J Microsc Spectrosc Electron 9:99–100 Pouchou JL, Pichoir L (1985) Les elements très légersen microanalyse X – Possibilités des modèles récents de quantification. J Microsc Spectrosc Electron 11:229–250 Procelli E (2004) Una facies a cavallo dello stretto, Rodì –TindariVallelunga e i rapporti tra Sicilia e Calabria nell’età del Bronzo. Atti XXXXVII Riunione scientifica Istituto Italiano di Preistoria e Protostoria, Firenze, pp 381–392 Renzulli A, Santi P, Nappi G, Luni M, Vitali D (2002a) Provenance and trade of volcanic rock millstones from Etruscan-Celtic and Roman archaeological sites in Central Italy. Eur J Mineral 14: 175–183 Renzulli A, Santi P, Serri G, Luni M (2002b) The Euganean trachyte flagstones (Bbasoli^) used by the Romans along the mid-Adriatic coast (Marche, central Italy): an archaeometric study. Periodico di Mineralogia 71:189–201 Roux V (1986) Le matériel de broyage à Tichitt (Mauritanie), Étude ethno-archéologique. Recherches sur les Civilisations, Paris Santi P, Antonelli F, Renzulli A, Pensabene P (2004) Leucite phonolite millstones from the Orvieto production centre: new data and insights into the Roman trade. Periodico di Mineralogia 73:57–69 Santi P, Renzulli A, Gullo R (2013) Archaeometric study of the hopperrubber and rotary Morgantina-type volcanic millstones of the Greek and Roman periods found in the Aeolian archipelago (southern Italy). Eur J Mineral 25:39–52 Santi P, Renzulli A, Bell M (2015) The volcanic millstones from the archaeological site of Morgantina (Sicily): provenance and evolution of the milling techniques in the Mediterranean area. Archaeometry 57:803–821 Storck J and Teague WD (1952) BHistory of milling^ From Canadian International Grain Institute (Minnesota University Press, 1952, via Copied Nisshin Flour Milling Inc Tigano G (2009) Milay II, Regione Siciliana, Assessorato Beni Culturali e Ambientali, Soprintendenza BCA di Messina. Servizio per i Beni Archeologici, Sicania-Messina Tigano G (2011) L’Antiquarium archeologico di Milazzo, Regione Siciliana, Assessorato Beni Culturali e Ambientali, Soprintendenza BCA di Messina. Servizio per i Beni Archeologici, Sicania-Messina Tusa S (1992) La Sicilia nella preistoria. II edizione, Sellerio, Palermo Williams D, Peacock D (2011) Bread for the people: the archaeology of mills and milling. University of Southampton, Series in Archaeology 3. Archaeo press, Oxford Williams-Thorpe O (1988) Provenancing and archaeology of Roman millstones from the Mediterranean area. J Archaeol Sci 15:253–305 Williams-Thorpe O, Thorpe RS (1990) Millstone provenancing used in tracing the route of a fourth century BC Greek merchant ship. Archaeometry 32:115–137 Williams-Thorpe O, Thorpe RS (1993) Geochemistry and trade of eastern Mediterranean millstones from the Neolithic to Roman periods. J Arch Sci 20:263–320 Williams-Thorpe O, Thorpe RS, Elliott C, Xenophontos C (1991) Archaeology, geochemistry, and trade of Igneous Rock millstones in Cyprus during the late bronze age to Roman periods. Geoarchaeology 6(1):27–60