(Re)sources: Origins of metals in Late Period Egypt (2018)

Journal of Archaeological Science: Reports 21 (2018) 318–339 Contents lists available at ScienceDirect Journal of Archaeological Science: Reports journal homepage: www.elsevier.com/locate/jasrep (Re)sources: Origins of metals in Late Period Egypt a,⁎ b c c Aurélia Masson-Berghoff , Ernst Pernicka , Duncan Hook , Andrew Meek a b c T The Department of Greece and Rome, British Museum, London, WC1B 3DG, UK Curt-Engelhorn-Zentrum Archäometrie gGmbH an der Universität Heidelberg, Mannheim, Germany The Department of Scientific Research, British Museum, London, WC1B 3DG, UK A R T I C LE I N FO A B S T R A C T Keywords: Late Period Egypt Metal Faience Lead isotopes Chemical analysis Provenance Metal trade and access to raw materials during the Late Bronze Age—roughly covering the New Kingdom in Egypt—have received substantial attention from past and present scholarship. Despite copper and lead remaining essential commodities afterwards, our knowledge about their supply during the Iron Age and later periods, in contrast, remains limited, even if it has improved recently. This paper presents the results of a pilot project investigating the possible sources of lead and copper available to Egypt during the Late Period (664–332 BCE), a period of intense contact and exchange between Egypt and the Mediterranean world. In the context of this research, a wide range of artefacts from Naukratis, a major cosmopolitan trading hub in the Western Nile Delta founded in the late 7th century BC, were analysed to determine their chemical composition and lead isotope ratios. They mostly consist of metal finds—including a crucible slag—but also some locally produced faience objects which used lead and copper to colour the glaze. Additional samples include metal objects and lead ores from Tell Dafana, a Late Period settlement in the Eastern Delta, and contemporary Egyptian or Egyptianizing bronzes from Cyprus. A total of 39 objects were analysed with a combination of lead isotope and elemental analysis, yielding surprising results for the likely origins of the copper. While Cyprus, an expected source for copper, is identified for one object, the copper deposits from Faynan or from northwestern Anatolia offer the best match for most finds, including those found in Cyprus. The lead analysed seems to originate from a variety of mines, particularly from Laurion in Attica, and mines in the northern Aegean and/or northwestern Anatolia, with one example possibly from a lead‑silver mine located in central Iran. The multiplicity of lead sources reflects the complexity of international trade in the Eastern Mediterranean at the time. The study offers a valuable insight into the trade networks of Egypt and, by extension, the whole of the ancient Mediterranean. A larger-scale project investigating objects from a wider range of sites in the Eastern Mediterranean world could revolutionize our understanding of metal trade and concomitant economic, political and social developments in the first millennium BC. 1. Introduction Unlike for the Late Bronze Age (LBA), textual and archaeological evidence is scarce when it comes to the trade and distribution of metals during the Iron Age and later periods (for a general survey of Iron Age evidence: Kassianidou, 2012; for additional evidence and discussion concerning the metal trade in Egypt in the later periods: Masson, 2015a). Lead isotope analyses (LIA), however, offer an appropriate means of discussing the origin of copper and lead ores, since a good deal of comparative data from ore deposits and raw copper and lead of the eastern Mediterranean and the Near East are now available. Few analyses so far have tried to address the question of the provenance of the ores of metal objects in Egypt dated to the Late Period (664–332 BCE), a period roughly covering the Archaic and Classical Greek periods that was characterized by a steep rise in the production especially of copper alloy statuettes for votive and ritual purposes (Roeder, 1937, 1956; Ogden, 2000; Hill and Schorsch, 2007; Weiss, 2012). Due to their particular deposition context, these statuettes proliferate in the material record in Egypt, but other types of copper alloy objects (weapons, vessels, furniture elements…) were also produced in large quantity during this period in the Mediterranean world (van Alfen, 2002). A sharp increase of the production of copper and that of Corresponding author. E-mail addresses: Amasson-berghoff@britishmuseum.org (A. Masson-Berghoff), [email protected] (E. Pernicka), [email protected] (D. Hook), [email protected] (A. Meek). ⁎ https://doi.org/10.1016/j.jasrep.2018.07.010 Received 12 March 2018; Received in revised form 25 June 2018; Accepted 12 July 2018 Available online 06 August 2018 2352-409X/ © 2018 Elsevier Ltd. All rights reserved. Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. lead (as a by-product of silver) in the Mediterranean world in the Archaic period was also measured by atmospheric pollution by copper and lead (de Callataÿ, 2005). An investigation, conducted more than thirty years ago, analysed 16 finds from Kawa and Sanam — Upper Nubian sites in modern Sudan — dated to the 25th dynasty (760–656 BCE), alongside four Late Period finds from Memphis, the ancient capital of Egypt (Fleming and Crowfoot-Payne, 1979; Fleming, 1982). A more recent study on numerous Late Period leaded bronze statuettes (Schulze and Lehmann, 2014) presents some methodological problems and resulted in unsupported interpretations (already discussed in Schwab and Willer, 2016). Finally, the results of LIA carried out on ten leaded bronze statuettes from Qubbet el-Hawa (Schwab and Willer, 2016), the necropolis of Elephantine (modern Aswan, in southern Egypt), and on eight unprovenanced lead curse tablets bought in Egypt and certainly originated in that country (Vogl et al., 2016) have recently been published and offer some insight into lead import to Egypt. No scientific investigation has yet addressed the provenance of the copper and lead ores used in the glazing of the Late Period faience despite its massproduction at the time (Guichard and Pierrat-Bonnefois, 2005). This is in stark contrast with the situation for LBA objects, on which numerous LIA have been carried out, both on copper alloy finds from Egypt and on copper ingots from the Mediterranean region (e.g. Stos-Gale et al., 1995; Gale and Stos-Gale, 2000; Begemann et al., 2001; Rademakers et al., 2017), as well as on faience and glass artefacts (Shortland, 2006). This paper represents a first step in addressing the current gap in knowledge and opening up new avenues of research, investigating the origin of copper and lead used in Late Period Egypt. It presents the results of chemical and LIA carried out on 39 objects. They comprise a wide array of Late Period finds (Table 1), including copper alloy statuettes, faience objects, lead ores and a crucible slag. The results are discussed in the context of a critical review of previous research and their wider historical implications are considered. The core of this investigation is a corpus of 31 metal and faience objects found at Naukratis, an Egyptian-Greek trading port founded in the 26th dynasty (664–525 BCE), more specifically in the final third of the 7th century BC. This harbour town of the western Nile Delta was strategically located on the Canopic branch, the most navigable branch of the Nile during the Late Period, between the seaport ThonisHeracleion guarding its Mediterranean entry point and Memphis at the apex of the Delta (Möller, 2000; Villing and Schlotzhauer, 2006; Demetriou, 2012, 105–152; Villing, 2015; Villing et al., 2013–19) (Figs. 1 and 2). It probably functioned as the international port for Sais, the 26th dynasty capital of Egypt. As such, Naukratis formed a major bridge between Egypt and many countries across the Mediterranean region. These cross-cultural and economic connections are reflected by the wealth of imports discovered at the site, originating particularly from Eastern Greece, Cyprus and the Levant. Egyptian, and particularly Lower Egyptian, material culture is nonetheless predominant at the site and Naukratis should not be seen as a simple Greek venture in Pharaonic land (Villing et al., 2013–19). By Pharaonic decree, taxes on ‘all goods that appear at Naukratis’ were due to the Royal Treasury during parts or potentially all of the Late Period (Agut-Labordère, 2012; von Bomhard, 2015). Since Naukratis was a major entry point for imports into Egypt and departure point for exports into the Mediterranean world during the Late Period, alongside Thonis-Heracleion and prior to the foundation of the seaport of Alexandria in 331 BCE, the material from this site is ideally suited for this type of investigation. Naukratis was not only a place of consumption and redistribution of imports, but also a centre of production of a variety of objects, meant for local, regional and/or international markets. Numerous workshops were active from the Late Period to the Roman period, with production of metal, faience, pottery, terracotta and possibly glass objects documented at the site. Petrie identified important metallurgic activities or groups of metal finds in Naukratis. On his plan of the town (Petrie, 1886, pl. XLV; Masson, 2015a, 84), he indicated several areas where he discovered evidence for silver working and copper ‘smelting’, as well as significant finds of iron tools, iron slags and iron ore (Fig. 2). Petrie also specified that the latter originated from the ‘low strata of the town’ which includes the stratum of the ‘scarab factory’, a workshop where scarabs and other faience amulets were produced in the first half of the 6th century BC (Masson, 2018a). Although Petrie went as far as defining Naukratis as a ‘great centre of the iron trade’ (Petrie, 1886, 39), he neglected to say anything of the size and significance of the ‘copper smelting’ activities he recorded on his map. It seems, however, dubious that smelting operations were undertaken at Naukratis, a site located far from any sources of raw materials. Nonetheless, there is a high probability that fresh supplies of metal would have been used in the various workshops active at the site given how vibrant an international hub it was – all the more so as large quantities of raw metals — tin, iron and copper — are recorded as having been imported into Egypt in the 5th century BCE in customs accounts found at Elephantine (discussed with references in Masson, 2015a, 79). The analysed objects were discovered during the late 19th and early 20th century explorations of the site conducted by W. M. Flinders Petrie, Ernest Gardner and David Hogarth, and are today kept in the British Museum, the Ashmolean Museum, Oxford, and the Petrie Museum, University College London (Figs. 3 and 4). The necessary historical and chronological framework for the wider assessment of the results was facilitated by the fact that this study is embedded in the British Museum's Naukratis Project. This project has not only been re-analysing the findings from the early explorations (Villing et al., 2013–19), but has also been investigating the site again with survey and excavations since 2012 (Thomas, 2015a). From the early excavations were collected at least 1118 objects in copper alloy and 42 finds in lead. However, the majority of these finds are difficult to date with accuracy. As Table 1 reveals clearly, some of the finds selected for analyses are insufficiently dated to conform exclusively to the Late Period (formally ending 332 BCE). This is due to the lack of specific context associated with the persistence of some types of objects into the early Ptolemaic period. The general interpretations and conclusions on metal trade for Late Period Egypt are therefore preliminary and, when it comes to the ill-dated finds, tentative. To provide some comparative data for the Naukratis assemblage, four samples were taken from finds discovered at Tell Dafana (Fig. 5), an eastern Nile Delta settlement excavated by Petrie (Petrie, 1888). The dating and context of the finds, kept at the British Museum, have recently been reassessed (Leclère and Spencer, 2014). Metallurgical activities involving copper and iron were observed at the site, with the evidence pointing more towards metalworking rather than smelting (Craddock in Leclère and Spencer, 2014, 142–143). From that site two lead ores were selected, for which we had no example from Naukratis, and two metal objects, all belonging to the 26th dynasty. Four samples were taken from objects found on Cyprus (Fig. 6). They include one mirror of Cypro-Achaic date (which can be compared with a mirror found at Naukratis that we analysed within the broader framework of this project), one ritual instrument (sistrum) probably made in Egypt and dated to the Late Period, and, two Egyptianizing statuettes dated to the Cypro-Archaic period. 2. Sample selection and archaeological assessment Selecting the most comprehensive range of material containing copper and/or lead was essential as it increases the probability of revealing distinct chemical compositions and metal origins. The following archaeological overview of the analysed finds, in metal and faience, includes a precise re-contextualising whenever possible. 2.1. Metal Beside the slag which results from local metallurgical activities (Fig. 3, cat. no. 1), Egyptian votive or ritual bronzes found at Naukratis (Fig. 3, cat. nos. 2–13) were probably produced locally or at least in the Delta (on Egyptian bronzes from Naukratis: Weiss, 2012, 442–446; Masson, 2015b). Some of them come from a cache of bronzes which produced 145 Egyptian bronzes according to Petrie who excavated it in 1885. From Petrie's brief description, it was possible to identify some of 319 320 Object Material Museum number Lab number Site Specific findspot Dating Fig. no. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Slag Neith statuette Ichneumon figure Eel votive box Mahes statuette Cobra votive box Osiris statuette Isis nursing statuette Wig, figure-fitting Bull statuette Atum statuette Situla Situla (base) Weight Weight Arrow-head Arrow-head Arrow-head Arrow-head Arrow-head Arrow-head Stamp-seal Mirror Bowl Bottle Dowel Lead ore Lead ore Stamp-seal Plaque Mirror Sistrum Worshipper statuette Worshipper statuette New Year's flask Scarab Scarab Scarab Scarab leaded tin bronze tin bronze leaded tin bronze leaded tin bronze arsenical copper leaded tin bronze leaded tin bronze leaded tin bronze tin bronze leaded tin bronze tin bronze leaded tin bronze leaded tin bronze tin-antimony bronze leaded antimony bronze leaded tin bronze leaded tin bronze leaded tin bronze leaded tin bronze tin bronze copper leaded tin bronze tin bronze tin bronze lead lead galena galena tin bronze tin bronze tin bronze tin bronze tin bronze leaded tin bronze faience faience faience faience faience London, Petrie Museum UC54639 BM EA27577 BM EA16040 BM EA27581 BM EA27594 BM EA27579 BM EA49132 BM EA49136 BM EA27599 BM EA27598 BM EA27597 BM EA27602 BM EA27587 Oxford, Ashmolean Museum AN1896-1908-E.3818 Oxford, Ashmolean Museum AN1950.370 BM 1886,0401.1739 BM 1888,0601.6.a BM 1886,0401.1736 BM 1886,0401.1737 BM 1935,0823.76 BM EA27509 BM 1886,0401.1706 BM 1886,0401.1743 BM 1886,0401.1746 BM 1888,0601.727 BM 1886,0401.31 BM EA23860 BM EA23556,l BM EA23903 BM EA23556,d BM 1894,1101.242 BM 1888,1115.19 BM 1873,0320.345 BM 1873,0320.340 BM EA58327 BM EA23617 BM EA66454 BM EA66477 BM EA66486 MA-154824 MA-154850 MA-154849 MA-154851 MA-154848 MA-154852 MA-154841 MA-154842 MA-154843 MA-154844 MA-154845 MA-154846 MA-154847 MA-154822 MA-154823 MA-154831 MA-154832 MA-154834 MA-154836 MA-154837 MA-154853 MA-154828 MA-154829 MA-154830 MA-154838 MA-154826 MA-154854 MA-154855 MA-154857 MA-154856 MA-154840 MA-154839 MA-154859 MA-154858 MA-155408 MA-155409 MA-155410 MA-155411 MA-155412 Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Naukratis Tell Dafana Tell Dafana Tell Dafana Tell Dafana Amathus (Cyprus) Palaepaphos (Cyprus) Idalion (Cyprus) Idalion (Cyprus) Naukratis Naukratis Naukratis Naukratis Naukratis Town (workshop) Cache of bronzes Cache of bronzes Cache of bronzes Cache of bronzes N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A Sanctuary of Apollo Sanctuary of Apollo? Sanctuary of Apollo Sanctuary of Apollo N/A N/A Town (house) N/A N/A Cemetery Sanctuary of Apollo Saite enclosure Foundation deposit Casemate building Foundation deposit Tomb 84 Sanctuary of Aphrodite N/A Sanctuary of Reshef-Apollo N/A N/A (made in Scarab Factory) N/A (made in Scarab Factory) N/A (made in Scarab Factory) Scarab Factory 600BC-500BC 500BC-350BC 500BC-350BC 500BC-350BC 500BC-350BC 600BC-100BC 630BC-332BC 630BC-200BC 630BC-332BC 630BC-200BC 630BC-332BC 630BC-332BC 630BC-200BC 630BC-300BC 630BC-300BC 630BC-525BC 525BC-30BC 525BC-30BC 630BC-400BC 630BC-450BC 630BC-525BC 500BC-400BC 630BC-350BC 630BC-600BC 400BC-200BC 575BC-560BC 664BC-525BC 664BC-610BC 570BC-526BC 664BC-610BC 600BC-300BC 750BC-500BC 650BC-500BC 650BC-500BC 600BC-525BC 600BC-570BC 600BC-570BC 600BC-570BC 600BC-570BC 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 5 5 5 5 6 6 6 6 4 4 4 4 4 Journal of Archaeological Science: Reports 21 (2018) 318–339 Cat no. A. Masson-Berghoff et al. Table 1 List of objects analysed with material, laboratory and museum inventory numbers, provenance (site and findspot), dating (mostly stylistic but also contextual and by inscriptions) and corresponding illustration number. Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. Fig. 1. Maps of the eastern Mediterranean and Western Asia with sites mentioned in the text, including cited Egyptian and Sudanese sites. 8 g, is also attested and could correspond to other standards such as the Near Eastern shekel or the Greek stater (Cour-Marty, 1990, 24–25, Figs. 22–24; Kroll, 2003, 317). The first analysed specimen, of 46 g, corresponds to a 5 kedets weight (cat. no. 14). The second, of 85 g, could also relate to an Egyptian standard, a 10 kedets weight but with a lighter unit of 8.5 g (cat. no. 15). Arrowheads represent another significant category of metal finds from Naukratis, with 88 inventoried specimens in total (Thomas, 2017). They belong to five broadly defined chronological groups ranging from the 7th century BC to the 1st century AD. Only the group represented by Roman tanged and barbed trefoil iron arrow-heads was excluded from the study. While some are characteristic of Egyptian types, and were most likely produced in Egypt, others could have been imported. Four out of the six analysed arrow heads were seemingly found in the sanctuary of Apollo. BM 1886,0401.1739 is a leaf-shaped arrow-head with a mid-rib and ending with a socket (Fig. 3, cat. no. 16) a type dated to between 630 and 525 BCE. The form developed out of New Kingdom types and is typical of Egyptian-style arrow-heads with known parallels dated to the late 7th to 6th centuries BC (Oren, 1984, 25, Fig. 26.2, 45). BM 1888,0601.6.a and BM 1886,0401.1736 are heavy barbed and tanged arrow-heads with a diamond section: the first (Fig. 3, cat. no. 17) belongs to the Dornpfeilspitzen type 1A4 and the second (Fig. 3, cat. no. 18) to either Dornpfeilspitzen type 1A4 or 1A5 (Baitinger, 2001, 98, pl. 2 nos. 34–35). These types — associated with composite bows — developed out of Late Bronze Age Greek and Cretan types, and were in use between the Archaic and Hellenistic periods (Erdmann, 1973, 35 Form B3, 1200, 7th to 1st centuries BC). The variants of these types found in Naukratis would have been used between 525 and 30 BCE (Thomas, 2017). BM 1886,0401.1737 is a leaf-shaped, socketed and barbed type that seems to originate in Anatolia and the Ionian cities of western Turkey (Fig. 3, cat. no. 19). This example has good parallels from the mid-7th to 5th centuries BC (e.g. Baitinger, 2001, Figs. 98–111, Zweiflügelige Tüllenpfeilspitzen type IIA2; Oren, 1984, 25, Fig. 26.3). For two further analysed arrow heads no specific find context is known. BM 1935,0823.76 (Fig. 3, cat. no. 20) is an arrow head known as ‘Scythian type’, with a short leaf shape, trefoil section and long socket. Socketed trilobate arrow-head types are common in the Middle East from the 7th century BC and subsequently in Greece and Egypt. They were particularly used by the Achaemenid Persians and Greeks (Baitinger, 2001, Fig. 284, Dreiflügelige Tüllenpfeilspitzen the finds, especially the bronzes of higher quality. A recent reassessment of its contents allowed this context to be dated to the late 5th–early 4th century BC (Masson, 2015a). However, since a cache is a secondary deposit, the dates of individual items can be quite heterogeneous, and some of the bronzes could date back to the late 7th–6th century BC (Davies, 2007, 183–184). In any case, a general Late Period date fits that group of objects. The other analysed Egyptian bronzes are generally of lesser quality and much harder to date, no more precisely than between the late 7th and the 3rd century BC, possibly even as late as the 2nd century BC and lack precise information on their find locations. They might be part of the cache of bronzes, such as the small cobra votive box (cat. no. 4): Petrie listed in his diary no less than 98 ‘bronze boxes’ from that cache, primarily for lizards, snakes and eels, the largest of which once used to contain the mummified remains of the animal they represent; however, similar votive boxes were uncovered in other contexts in Naukratis and these too are difficult to date specifically (Masson, 2015a, 79). The gods of the Osirian triad, Osiris, his sister-wife Isis and their son Horus, prevail in the corpus of bronze statuettes from Naukratis, as in many Late Period sites throughout Egypt (Weiss, 2012). Osiris alone is represented in the cache of bronzes with 19 statuettes, but also at least further 13 statuettes were discovered elsewhere in Naukratis (Masson, 2015a, 77). Other analysed copper alloy objects could have been produced locally while others could have been imported. The latter case could apply to the two weights we analysed (Fig. 3, cat. nos. 14–15). Weights form a very important category of finds preserved from the early excavations at Naukratis, with over 1000 extant examples. They belong to Egyptian, Greek and Near Eastern standards and exemplify the thriving trade and exchange between Egypt and the Near Eastern and Mediterranean worlds that passed through Naukratis (Petrie, 1886). Although they are predominantly made out of different types of stone, 25% of this rich and varied assemblage is made of metal (Masson, in preparation). The weights selected for analysis are of truncated conical shape, with flat base and convex top. Such dome-shaped weights are the most common type found in Naukratis (Petrie, 1886, pl. XXIII, ‘domed’ type), but also elsewhere in Egypt (Cour-Marty, 1990, 25–26, Fig. 4, Fig. 13) and in the Near East (Birney and Levine, 2011, 474). Dome-shaped weights are especially associated with the Egyptian standard deben-kedet (usually around 90–95 g for the deben and 9–9.5 g for the kedet) and their multiples or sub-multiples (Cour-Marty, 1990, 20, 25, Figs. 22–24). However a lighter unit, around 321 Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. Fig. 2. General map of Naukratis (incorporating all fieldwork by Petrie, Gardner, Hogarth, Coulson, Leonard, Thomas and Villing and preliminary geophysics results). Location of major metal finds and metallurgical activities following Petrie's indication (1886, pl. XLV). Map by Ross Thomas © Trustees of the British Museum. 322 Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. Fig. 3. Illustrations of metal finds from Naukratis analysed in this study (listed in Table 1; not to scale). 1): Photograph © Petrie Museum of Egyptian Archaeology, UCL. 14) and 15): Photographs © Ashmolean Museum, University of Oxford. All other images: Photographs © Trustees of the British Museum. to an Egyptian form not found in Greece or the Middle East, but has been found in Israel, notably at Ashkelon in 7th century BC contexts (Aja, 2011, 536, nos. 78–82) and this type is replaced with the introduction of Scythian types during the Persian rule of Egypt in the 6th century BC, discussed above. It is likely an early form that was abandoned early in Naukratis' history (Thomas, 2017). type IIB3). The long socketed variant broadly dates to the period between 630 and 450 BCE, though it is less common in the 5th century BC and absent at Marathon (Erdmann, 1973), by which time it had been replaced by shaftless short socket types. The leaf-shaped arrow-head BM EA27509 (Fig. 3, cat. no. 21) has a central rib visible beneath corrosion on both sides. This oblanceolate tanged arrow-head belongs Fig. 4. Illustrations of faience finds from Naukratis analysed in this study (listed in Table 1; not to scale). Photographs © Trustees of the British Museum. 323 Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. (525–404 BCE). It was discovered in a house located in the southern part of Naukratis (Petrie, 1886, 41, pl. XX, no. 17). Seals of this type were usually used to stamp mud or plaster jar-sealings. Several such sealings as well as a seal in bronze, featuring the plumed cartouche of pharaohs of the 26th dynasty (664–525 BCE), were discovered in Tell Dafana (Petrie, 1888, 77 and 111, pl. XLI no. 76). We analysed the bronze seal to provide a comparison for the Naukratis seal. The rectangular seal has a plumed cartouche with the name of the pharaoh Amasis (570–526 BCE) preceded and followed by the epithets ‘Perfect God’ and ‘Son of Neith’ (Leclère and Spencer, 2014, 66, pl. 24). This second seal is doubtlessly an Egyptian product (Fig. 5, cat. no. 29). The solid-cast mirror BM 1886,0401.1743 is one of nine bronze mirrors discovered during the early exploration of Naukratis (Fig. 3, cat. no. 23). Its circular shape with a simple stepped projecting tang, to which was once attached a now missing wooden, bone or ivory handle, is characteristic of the Late Period (Thomas and Acosta, 2018). Mirrors had more than a purely domestic function. According to iconographic and archaeological evidence, they could be an object of cult, either offered as an ex-voto or part of the funerary equipment (Robinson, 2010, 219–220). This mirror is heavy (957 g), possibly indicative of its votive function (e.g. the bulk of mirrors found at Thonis-Heracleion weights ca. 200 g: Robinson, 2010, 219). Mirrors of similar shape were discovered in Egypt, for example at Thonis-Heracleion, where the shape has been identified as typically Egyptian (Robinson, 2016, 112). This mirror is plain, but another example from Naukratis is decorated with volutes (Oriental Institute, Chicago, inv. E18836). Whether plain or decorated with volutes, these types of mirror were also commonly found in Cyprus, such as BM 1894,1101.242, a mirror from the bronzerich Tomb 84 of Cypro-Archaic date in Amathus (Murray et al., 1900; for the type: Gjerstad, 1948, 142–143, type 2 and Fig. 25, 2; Chavanne, 1990, 12–13, pls 4–5 and 21), that was chosen for analysis (Fig. 6, cat. no. 31). These mirrors could have been produced in Egypt, Cyprus, or even in the Levant (see Gjerstad, 1948, 381 where the type is derived from North Syria based on iconographic evidence). Metal vessels are rarely preserved at Naukratis. Among the few that survived are a bronze hemispherical bowl (Fig. 3, cat. no. 24) and a lead bottle (Fig. 3, cat. no. 25). The handleless bowl with a pronounced inner lip could be a direct import from Cyprus (Petrie, 1886, pl. 12) since many similar bowls were deposited in Late Cypriote III, CyproGeometric and Cypro-Archaic tombs (Matthäus, 1985, 71–104; Chavanne, 1990, 1 note 14, pls 1 and 20; for more recent finds, see for example from Palaepaphos: Karageorghis and Raptou, 2014, 74, pls XXXIX and XCI, nos. 62–64 and 71). The lead bottle is later in date. The earliest known lead vessels in the Mediterranean world are dated to the 6th century BC, though it is not before the end of the Late Period – beginning of the Ptolemaic period that we get them in significant numbers notably in Egypt (Gubel and Cauet, 1987; van der Wilt, 2014). The bottle was uncovered in the cemetery of Naukratis, where the majority (though certainly not all) artefacts date to the 4th and 3rd century BC (Gardner, 1888, 28; Villing, 2015, 234–236) and could therefore belong to that timeframe. The lead dowel from a fragmentary limestone palmette, probably of an acroterion (Fig. 3, cat. no. 26) was possibly used as a mending piece, since a join and pin were found on the bottom surface. The shape of the palmette seems to be Archaic (c. 575–560 BCE) and suggests that it came from a small piece of architecture (altar, base or stele). Petrie had allocated this fragment to the first temple of Apollo (Petrie, 1886, pl. 14 A) but this is unlikely given its small size and date (Koenigs, 2007, no. 38, p. 342, pl. 26). To this list of metal finds, we added a small bronze plaque from the site of Tell Dafana (Fig. 5, cat. no. 30). It was discovered in the foundation deposit of the pharaoh Psamtik I (664–610 BCE) associated with the site's large Egyptian temple (Petrie, 1888, 54–55, pls XXII–XXIII; Leclère and Spencer, 2014, 54, pl. 17). Foundation deposits of the Late and Ptolemaic periods often contained several such uninscribed plaques and samples of a wide variety of materials (Weinstein, 1973; Masson, 2015c). From the same foundation deposit comes one of the two lead Fig. 5. Illustrations of finds from Tell Dafana analysed in this study (listed in Table 1; not to scale). Photographs © Trustees of the British Museum. Fig. 6. Illustrations of finds from Cyprus analysed in this study (listed in Table 1; not to scale). Photographs © Trustees of the British Museum. In addition to objects representing the major categories of metal finds at the site we have selected some artefacts which are unique or relatively rare as well as some comparative material from Tell Dafana and Cyprus. A bronze stamp-seal in a shape of a plumed cartouche (Fig. 3, cat. no. 22) bears an Aramaic inscription, possibly the name of an official (Villing, 2013a, 75, Fig. 1; Masson, 2018b). It can be dated to the period of Persian domination in Egypt, during the 27th dynasty 324 Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. Table 2 Chemical composition of the samples analysed with EDXRF, normalized to 100%. All values are given in mass percent. Selenium was below 0.01% and tellurium as well as gold below 0.005% in all samples. ‘n.d.’ means “not determined”. Figures in italics indicate partly corroded samplesa. Cat no. Object Museum number Lab no. Cu Fe Co Ni Zn As Ag Sn Sb Pb Bi 1 2 3 4 5 6 7 8 London, Petrie Museum UC54639 BM EA27577 BM EA16040 BM EA27581 BM EA27594 BM EA27579 BM EA49132 BM EA49136 MA-154824 MA-154850 MA-154849 MA-154851 MA-154848 MA-154852 MA-154841 MA-154842 50 87 82 88 93 81 78 81 1.11 0.05 0.81 < 0.05 0.59 0.55 < 0.05 < 0.05 0.03 0.12 0.03 0.02 0.02 0.04 0.05 0.05 0.02 0.05 0.04 < 0.01 0.02 0.05 0.05 0.09 0.3 < 0.1 < 0.1 < 0.1 0.5 < 0.1 0.5 0.3 0.43 0.2 0.6 0.3 3.2 1 0.1 < 0.1 0.02 0.021 0.022 0.035 0.109 0.022 0.028 0.033 3.2 9 7 4.8 0.38 2.53 3.1 3.5 0.13 0.01 0.05 0.03 0.34 0.3 0.03 0.05 45 3.8 9 7.3 2.2 15 18 15 < 0.01 0.01 0.1 0.02 < 0.01 0.06 < 0.03 < 0.04 9 10 11 12 13 14 Slag Neith statuette Ichneumon figure Eel votive box Mahes statuette Cobra votive box Osiris statuette Isis nursing statuette Wig, figure-fitting Bull statuette Atum statuette Situla Situla (base) Weight MA-154843 MA-154844 MA-154845 MA-154846 MA-154847 MA-154822 88 81 86 71 80 81 0.59 0.35 0.1 5.14 1.37 0.06 0.02 0.01 0.05 0.02 0.05 0.01 < 0.01 0.03 0.21 0.03 0.07 0.02 0.5 < 0.1 < 0.1 < 0.1 0.4 < 0.1 < 0.1 0.89 0.06 0.2 0.1 0.22 0.008 0.064 0.033 0.082 0.019 0.142 8.1 1.58 10.6 0.95 9.3 7.3 0.02 0.27 0.07 0.1 0.06 9.1 2.5 15 2.5 22 8.8 2.6 < 0.05 0.05 < 0.01 0.05 0.01 0.03 15 Weight MA-154823 71 0.18 0.18 0.12 < 0.1 0.09 0.037 0.02 10.3 18 0.05 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 Arrow-head Arrow-head Arrow-head Arrow-head Arrow-head Arrow-head Stamp-seal Mirror Bowl Bottle Dowel Lead ore Lead ore Stamp-seal Plaque Mirror Sistrum Worshipper statuette Worshipper statuette BM EA27599 BM EA27598 BM EA27597 BM EA27602 BM EA27587 Oxford, Ashmolean Museum AN1896–1908-E.3818 Oxford, Ashmolean Museum AN1950.370 BM 1886,0401.1739 BM 1888,0601.6.a BM 1886,0401.1736 BM 1886,0401.1737 BM 1935,0823.76 BM EA27509 BM 1886,0401.1706 BM 1886,0401.1743 BM 1886,0401.1746 BM 1888,0601.727 BM 1886,0401.31 BM EA23860 BM EA23556,l BM EA23903 BM EA23556,d BM 1894,1101.242 BM 1888,1115.19 BM 1873,0320.345 MA-154831 MA-154832 MA-154834 MA-154836 MA-154837 MA-154853 MA-154828 MA-154829 MA-154830 MA-154838 MA-154826 MA-154854 MA-154855 MA-154857 MA-154856 MA-154840 MA-154839 MA-154859 88 76 77 80 91 99 87 91 93 0.4 0.1 0.2 0.1 92 88 93 91 91 0.43 0.3 0.21 0.52 0.06 0.74 0.08 0.13 0.18 n.d. 0.05 n.d. n.d. 0.12 0.35 0.18 0.13 0.1 0.02 0.06 0.14 0.03 0.04 0.02 0.02 0.01 < 0.01 n.d. n.d. n.d. n.d. 0.02 0.14 0.02 0.01 0.02 0.02 0.03 0.02 0.02 < 0.01 0.02 0.05 0.01 0.01 n.d. n.d. n.d. n.d. 0.03 0.1 < 0.01 < 0.01 0.05 0.4 < 0.1 0.2 0.5 0.8 < 0.1 < 0.1 0.2 < 0.1 n.d. n.d. < 0.1 < 0.1 < 0.1 < 0.1 0.6 0.6 < 0.1 0.42 < 0.1 0.05 < 0.1 0.1 0.45 0.34 0.2 0.29 n.d. 0.06 0.31 0.14 0.07 0.28 0.14 0.29 1.19 0.023 0.017 0.011 < 0.002 0.012 0.003 0.036 0.024 0.008 0.016 0.026 0.035 0.011 0.019 < 0.002 0.016 0.008 0.011 2.9 3.9 5.4 1.39 7 0.03 3.2 7.7 6.9 0.03 n.d. 0.01 n.d. 4 11.4 4.9 7.6 7.2 0.04 0.03 0.05 0.61 0.04 0.01 0.1 0.05 0.03 0.05 n.d. n.d. n.d. 0.45 < 0.005 0.04 0.03 0.03 7.7 20 17 17 1.07 0.11 9.4 1.17 0.02 99 100 99 100 3.1 0.04 0.82 0.03 0.18 < 0.03 0.04 < 0.01 < 0.01 0.1 0.01 0.05 < 0.01 < 0.02 n.d. 0.02 n.d. n.d. < 0.01 < 0.01 < 0.02 < 0.01 < 0.01 BM 1873,0320.340 MA-154858 86 2.12 0.03 0.02 < 0.1 0.05 0.016 3.2 0.19 8.5 0.01 34 a In some corroded samples the tin content could be enhanced; however the tin contents are not systematically higher than in the uncorroded metal samples. ore samples we analysed (Fig. 5, cat. no. 28). Three votive bronzes found in Cyprus were added to the sample set so as to compare them with the Egyptian votive or ritual bronzes from Naukratis. One is a Hathoric sistrum (BM 1888,1115.19) discovered in Palaepaphos that was most likely deposited in the local sanctuary of the Great Goddess/Aphrodite (Fig. 6, cat. no. 32); on stylistic grounds it has been identified as a 26th dynasty Egyptian import rather than a local or Phoenician Egyptianizing product; as such, this represents a rare import of a genuine Egyptian product (Carbillet, 2011, no. C3). The two others are Cypro-Archaic statuettes, BM 1873,0320.340 and BM 1873,0320.345 (Fig. 6, cat. nos. 33–34), both from Idalion, with the first deposited in the sanctuary of Reshef-Apollo (Masson, 1968, 393–396; Reyes, 1992, cat. no. 15, p. 247, pl. 15c; for the sanctuary, see Senff, 1993, esp. 5–12 for layout and findspots). They are Egyptianizing statuettes of worshippers wearing an Egyptian-style costume and collar which were dedicated alongside limestone examples of a similar Egyptianizing style (Senff, 1993, 50–53; Faegersten, 2003). These cannot be interpreted as actual Egyptian products and were produced in Cypriot or Phoenician workshops. Iron Age. Naukratis is again a good starting point for such an investigation since it was a major centre for the production of faience objects that were widely exported across the whole Mediterranean region, as far as the Levant, Tunisia, southern Russia, Greece, Italy, Spain, Libya and Cyprus (Gorton, 1996, 91–131; Masson, 2018a). In 1885, Petrie uncovered the discarded waste of a workshop in the vicinity of the sanctuary of Aphrodite (Petrie, 1886, 35–37) (Fig. 2). This workshop, known as the ‘Scarab Factory’, was active mainly between 600 and 570 BCE. It specialized in the mass production of amulets, primarily in the form of scarab beetles in faience and ‘Egyptian blue’. Amulets, hundreds of associated moulds and some raw materials used in the production were discovered in the rubble (Masson, 2018a). Four mould-made scarabs from the ‘Scarab Factory’ and one New Year's flask were sampled for this study. Although few examples of New Year's flasks have survived from the early excavations (on this type of objects: Caubet and Pierrat-Bonnefois, 2005, 148), we know from the archaeologists' diaries and publications that they were a common find category at the site and were likely produced there (Masson, 2014). New Year's flasks are particularly characteristic of the 26th dynasty and they still persist in some 27th dynasty contexts (525–404 BCE). A 6th century BC date seems probable for the Naukratis examples (Masson, 2014). The green glaze on the fragmentary specimen we analysed is quite worn (Fig. 4, cat. no. 35). Each selected scarab bears a motif typical of the products from the Scarab Factory (Fig. 4, cat. nos. 36–39) (Gorton, 1996, cat. nos. A17, A50, A52 and B97). BM EA23617 retains its glossy yellow glaze. BM EA66454 features a pale greenish turquoise glaze relatively well 2.2. Faience A study conducted on a wide range of Late Bronze Age Egyptian products has demonstrated that while Egyptian lead ores were not used for smelting lead, they served other applications, such as the manufacturing of faience objects (Shortland, 2006). The aim of this pilot study was to investigate if such a phenomenon persisted into the Late 325 Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. without reference material, much better results are obtained by calibration with reference materials of known compositions. In the calculations it is assumed that the measured elements sum up to 100%, which is largely fulfilled with ancient copper alloys. All samples were measured twice with different instrument settings (Lutz and Pernicka, 1996) to optimise for two groups of elements with atomic numbers 20–42 and 66–92 as well as those with atomic numbers 46–56. Detection limits are 0.01% for the first group of elements, 0.005% for the second group, and 0.1% for elements with interferences like zinc and arsenic in the presence of lead. Precision is better than 10% relative above 0.1%. For the two lead samples and the two lead ores no reference material was available for all elements considered for the copper-based alloys. However, Co, Ni, and Zn are neither expected nor informative in ancient lead metal anyway. Furthermore, the lead ores were measured as powders without determination of sulphur. Accordingly, the compositions listed in Table 2 are normalized to 100% which corresponds to the composition of the lead metal that results on smelting. Compositional data were already available for the five faience samples from non-destructive proton induced X-ray emission (PIXE) spectrometry under a 3 MeV energy proton beam for characterization of the elemental chemical composition (major, minor and trace elements) by the C2RMF (Centre de Recherche et de Restauration des Musées de France) (results published in Meek et al., 2016). They were, however, sampled for the purposes of lead isotope analysis. Lead isotope analysis (LIA) of all 39 objects was accomplished by multiple-collector inductively-coupled plasma mass spectrometer (MCICP-MS) at the Curt-Engelhorn-Center for Archaeometry in Mannheim. Each sample was dissolved in diluted HNO3 and lead was separated with ion chromatography resin from the matrix. Details are described in Niederschlag et al. (2003). The isotope ratios of lead were corrected for the mass discrimination by addition of Tl. A value of as 205 Tl/203Tl = 2.3871 was taken and an exponential relationship assumed. 204Pb was corrected for the isobaric interference with 204Hg by measuring 202Hg and using a 204Hg/202Hg ratio of 0.2293. The in-run precision of the reported lead isotope measurements was in the range of 0.01 to 0.03% (2σ) depending on the ratio considered. Fig. 7. Tin and lead concentrations in the copper-based alloys of this study. There is no correlation between those two metals so they must have been introduced separately. The horizontal line at 4% indicates the limit between lead as impurity in the alloy and intentionally added lead. This differentiation is not very clear-cut but is suggested by the relatively large gap between 4 and 7% lead. It is also relevant for the discussion of the lead isotope ratios (see below). preserved on the underside, but the motif itself appears unglazed. This scarab might have had originally a bichrome glazing, with a yellow motif on a turquoise background. BM EA66477 possesses a dark yellow glaze, with some tinge of green, while BM EA66486 presents an overall pale green glaze. 3. Methodology The main aim of our study was to determine the provenance of the raw metals used in the production of the analysed objects. Although the trace element pattern does carry some information on the geological source, this set of parameters is often not sufficient to distinguish between several possible ore sources. Since the 1970s it has therefore become customary to use stable lead isotope ratios in addition to trace element concentrations for this purpose. The first applications were on lead and silver objects from the Aegean which could thus be related to specific ore sources (Gale et al., 1980). This combination of methods was later expanded to include copper-based alloys (Gale and Stos-Gale, 1982) and is nowadays widely employed in provenance studies (Pernicka, 2014). For this type of analysis it is highly recommended to extract a small sample from the uncorroded metal that has to be treated in the laboratory to separate the lead from the matrix in order to obtain the highest possible precision of measurement. Accordingly, in this study altogether 31 samples from metal objects — consisting of 29 copperbased and 2 lead objects — were obtained by drilling with a high speed steel drill of 1 mm diameter at low speed. In addition, two samples of lead ore from Tell Dafana and one sample of slag from Naukratis were included. Compositional analysis was carried out on these 34 samples to characterise their chemical composition. For routine analysis of metal samples an energy-dispersive X-ray spectrometer (ARL Quant'X by Thermo Scientific) was used at the CurtEngelhorn-Center for Archaeometry in Mannheim. It is equipped with an automatic sample changer and an X-ray tube with a rhodium anode. The characteristic secondary X-rays emitted by the sample are measured with an electrically cooled Si(Li) semiconductor detector. The spectra are deconvoluted and evaluated with the WinTrace-Software supplied by Thermo Scientific. Interelement effects are corrected with the fundamental parameter method (FP method, Criss et al., 1968). Although this method (ED-XRF) in principle allows the calculation of element concentrations 4. Results: chemical composition As the results concerning the chemical composition of the faience objects have already been published as a part of a wider study of glazed artefacts from Naukratis and Rhodes (Meek et al., 2016), only the new results for the chemical composition of ores and metal finds are presented in detail in the following (Table 2). For provenance studies of copper or copper alloys by lead isotope analysis one must assume that the lead in the copper or copper alloy derives from the copper deposits as an impurity and was not added intentionally. Tin ores and thus tin metal are usually very low in lead (Gmelins Handbuch der Anorganischen Chemie, 1971) so that the addition of tin to produce tin bronze would not alter the lead isotope ratios significantly. On the other hand, in cases where even small amounts of lead are added to an alloy, the lead isotope ratios will then indicate the provenance of the lead rather than that of the copper. It is difficult to decide with certainty, if lead was added intentionally or not. Most researchers tend to set an upper limit of ca. 2% lead in a copperbased alloy to be regarded as impurity. Above this level one has to deal with the possibility that lead was added intentionally. This can be considered as certain with lead concentrations above ca. 5%. In our sample suite three groups can be distinguished concerning the lead concentrations: one with lead below about 4%, one with lead between 7 and 10% and one group with high lead concentrations above 14%. For convenience, we regard the first group as unalloyed with lead despite the uncertainty between 2 and 4% lead (Fig. 7). It is clear from the compositional data, that there is no correlation between categories of objects and their chemical composition. Of the 29 samples from copper-based objects about half are below the above explained threshold of 4% lead (Fig. 7) and only these can be 326 Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. casting. It is typical that the lead is enriched in such crucible slags and, therefore, the composition is not representative of the composition of the metal that was melted (Rademakers et al., 2017). However, this would have to be checked by a mineralogical analysis that was not planned for this pilot project. discussed in terms of the provenance of copper. For the remaining samples the lead isotope ratios indicate the provenance of the lead rather than the copper. It is important to take this into consideration to avoid misleading conclusions as they were reached in a recent study of Egyptian bronze statuettes of the Late Period by Schulze and Lehmann (2014), who suggested Cyprus as most likely source of the copper although most of their analysed objects consisted of leaded bronze. Their use of pXRF is another potential problem (see below). Although Cyprus is certainly a possible source of copper for ancient Egypt, there are no lead deposits on the island and the copper deposits are characteristically low in lead (Constantinou, 1982; Gale et al., 1997; Charalambous et al., 2014), so that the lead isotope ratios in a leaded copper alloy can by no means be related to Cyprus. Leaded copper alloys were predominantly used in the Late Period in Egypt (Ogden, 2000). The addition of lead reduces the melting and solidification temperature of copper slightly and also reduces the viscosity of the melt and thus decisively improves the casting properties of the alloy (see below). Furthermore, lead was an exceedingly cheap commodity, because lead was a by-product of silver production that was of little use before the Hellenistic and especially the Roman period. Tin concentrations in the analysed objects can be divided somewhat arbitrarily into two groups, one with more than about 6% tin that can be regarded as traditional tin bronzes with a mean value of about 10% tin and a second one with less than 5% tin. Low-tin bronzes seem to be the rule rather than the exception in Late Period Egypt (Riederer, 1988; Schwab and Willer, 2016). However, bronze with less than about 2% is more difficult to cast, because copper can dissolve oxygen in the molten state and gives it off on solidification, which produces gas bubbles in the cast (Tafel and Wagenmann, 1951). Since tin also serves as an antioxidant, this tendency is reduced at higher tin concentrations. High lead concentration can partially substitute for tin in this respect but at the cost of a less attractive colour of the alloy which tends to become greyish. Only one object, an arrow-head from Naukratis (Fig. 3, cat. no. 21), consists of unalloyed copper. There are four objects that contain between 0.4 and 2% tin: another arrow-head (Fig. 3, cat. no. 19), the bull statuette (Fig. 3, cat. no. 10), a situla (Fig. 3, cat. no. 12) and the Mahes statuette (Fig. 3, cat. no. 5). Such low tin concentrations have practically no recognizable effect on the properties of the alloy and are usually considered as indication for recycling of metal (Figueiredo et al., 2010). This introduces a further complexity, because in this case it is even more uncertain, from which ore the lead isotope ratios derive. On the other hand, the composition of one of these objects, the Mahes statuette, has rather high concentrations of arsenic, antimony and iron, all elements that tend to be depleted on remelting or recycling. Accordingly, it is unlikely that this object was produced from recycled metal and the low tin concentration may be an impurity either from the ore or from the workshop. 4.2. Egyptian votive or ritual bronzes The composition of the Egyptian votive or ritual bronzes that can be attributed to the cache is quite disparate. The Neith's statuette (Fig. 3, cat. no. 2) consists of a true tin bronze with a little lead. The mummycase in the shape of an ichneumon (Fig. 3, cat. no. 3) and the large votive box topped by an eel figure (Fig. 3, cat. no. 4) are tin bronzes, with different contents of tin (7.0% for the mongoose, 4.8% for the eel votive box). Their lead concentrations (9.0% for the mongoose, 7.3% for the eel votive box) indicate intentional addition of lead. Finally, the hollow-cast statuette of the male lion-headed deity Mahes, which is of quite exceptional quality, is an arsenical copper (Fig. 3, cat. no. 5). Its high arsenic content (3.2%) may indicate that arsenic was added intentionally (e.g. by adding speiss, see Thornton et al., 2009; Rehren et al., 2012), rather than being a natural occurrence in the copper ore. Arsenical copper is rarely found after the New Kingdom (Ogden, 2000, 153). Analyses have revealed, however, that cat statuettes and catheaded deities are a group of Egyptian copper alloys with high arsenic contents (i.e. Riederer, 1988; Schorsch, 1988). A couple of bronzes depicting the child deity Harpokrates, produced after the New Kingdom, also presented high level of arsenic, perhaps as arsenical copper is lighter in colour and children were often shown with a paler skin than adults in Egyptian art (Ogden, 2000, 153). As a lion-headed deity and son of Bastet, it seems that Mahes fits well the profile of other objects of arsenical copper so far identified in the Late Period. The other analysed Egyptian bronzes that belong to mass-produced types of small-size and lesser quality show a different composition, more aligned with what we know for many Late Period votive bronzes (e.g. Wuttmann et al., 2008; Schwab and Willer, 2016). The small cobra votive box (Fig. 3, cat. no. 6) is a low-tin bronze (Sn: 2.53%) with a high level of lead (Pb: 15%). The Osiris statuette we analysed also has a high lead content (17.7%) (Fig. 3, cat. no. 7) as does the bronze statuette depicting Isis nursing Horus (14.7%) (Fig. 3, cat. no. 8). Both are lowtin bronzes with added lead and are chemically rather similar. Unfortunately, these analyses cannot reliably be compared with the results recently published by Schulze and Lehmann (2014), who employed non-destructive XRF analysis with a portable instrument (pXRF). This technique risks providing only the composition of a corroded or otherwise altered surface rather than the alloy's bulk composition (on problems with the use of pXRF in archaeometallurgy, see e.g. Nørgaard, 2017; cf. also Hall, 1961). This is especially true for tin bronze and leaded tin bronze, which are the dominant alloys in Late Period Egypt, as tin and lead form insoluble compounds on corrosion and tend to be substantially enriched on the surface. Conservation treatments — such as alkaline solutions containing metallic zinc that were long used to treat corroded metal objects in order to reduce the oxidic components in the corrosion layer, leading to enhanced zinc concentrations and thus mimic the presence of brass (e.g. early first millennium BC object identified as brass in Schulze and Lehmann, 2014, though brass is not attested in Egypt before the Roman period) — can further alter surface composition, making a use on well-cleaned surfaces imperative (Charalambous et al., 2014). More significant for comparison are the analyses by Riederer (1981, 1988), obtained from small samples from the interior of the objects by atomic absorption spectrometry. In a large number of Osiris statuettes Riederer detected lead concentrations ranging from a few percent up to almost 30% lead, combined with variable tin concentrations between zero and 12% tin. Accordingly, there is no indication that a particular alloy type was used for this type of statuettes. Statuettes recently analysed from a tomb at Qubbet el-Hawa mostly represent Osiris and another Isis (or Hathor). These 6th–5th century BC statuettes are part of an assemblage of artefacts related to 4.1. Ores and slag The two lead ore samples from Tell Dafana (Fig. 5, cat. nos. 27–28) consist of relatively pure galena (PbS) the most abundant lead ore that was also used as a pigment and for cosmetic purposes in Egypt (abundant literature quoted in Hallmann, 2009). Argentiferous galena is also the major ore for silver production in antiquity, but the ore samples from Tell Dafana contain too little silver to have served for this purpose. In the Late Bronze Age the economic threshold for the extraction of silver from lead seems to have been around 0.03% Ag and from the Roman period on around 0.01% Ag (Pernicka, 1987). Slags are important indicators of the type of production activities that took place. Three stages of production produce distinctive slags: reduction, purification, and elaboration (melting) (Pichot et al., 2006, 220–222; Heinz, 2015, 57). Contrary to what Petrie had suggested, it is highly unlikely that any copper smelting took place at Naukratis. This is also supported by the analysis of one of the few slags that survived from Petrie's fieldwork (all kept in the Petrie Museum). The composition of one sample (Fig. 3, cat. no. 1) suggests that it is a crucible slag produced during melting of a highly leaded low-tin bronze, presumably for 327 Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. the production of metal statuary, including wax models and failed castings (Auenmüller and Fitzenreiter, 2014; Auenmüller, 2014). Their lead content is again high — between 7.7 and 21% — while their tin content varies between 0.71 and 6.1% (Schwab and Willer, 2016). Cowell, who conducted a series of analyses (atomic absorption spectrometry and, in some cases, X-ray fluorescence spectrometry) on metal finds kept in the British Museum, suggested that a lead-rich alloy combines a greater fluidity with a rather low melting point, characteristics that would allow casting complex pieces (Davies and Cowell, 1987). As noted by Charalambous et al. (2014, 213), however, the addition of only 2 to 3% of lead could produce such an effect. As a byproduct of silver production, it is assumed that lead was a cheaper commodity, and the high level of lead in many Late Period Egyptian bronzes could be explained by the low cost of lead in comparison to copper and tin. It would have allowed the cheap production of countless votive and ritual statuettes, many of which are small, plain and of poor quality. For example, out of more than 1800 Egyptian bronzes discovered between 1964 and 1976 at the Sacred Animal Necropolis at North Saqqara, only a few displayed great craftsmanship while the majority were clearly mass-produced (Davies, 2007). Recycling of Egyptian ritual metal statuary is seldom identified (Fitzenreiter et al., 2016; Schorsch, forthcoming). The numerous 1st millennium BC deposits of dozens, hundreds or even thousands of metal statuettes preserved through (often ritual) burials, usually within or in proximity of religious complexes (e.g. Weiss, 2012) indicate that this type of material often withdraws metal from circulation into permanent deposition. However, it does not preclude the use of recycled metal in their production for at least some of them, as indicated by analyses carried out on a few Naukratis statuettes (cat. nos. 10 and 12, but unlikely no. 5). archer-god, it is probable that they were non-functional votive offerings. Dedicating arrows is a practice attested in other Greek sanctuaries, notably during the Classical period (Baitinger, 2001). It seems possible that some of these arrow heads in leaded bronze were produced locally in Egypt, perhaps even in Naukratis, but their import cannot be excluded, especially for nonEgyptian types. In stark contrast, two further arrow heads from Naukratis (Fig. 3, cat. nos. 20–21) for which no specific findspot was recorded, contain very low-level of lead, between 0.11% and 1.07%. These might have been used, or at least produced, as actual weapons. 4.5. Other metal finds Apart from the lead bottle (Fig. 3, cat. no. 25; Pb: 99%) and lead dowel (Fig. 3, cat. no. 26; Pb: 100%), the remaining analysed finds showed a range of chemical compositions. The two stamp-seals are low-tin bronzes with lead: the seal from Tell Dafana (Fig. 5, cat. no. 29) has only about 3% lead whereas the one from Naukratis (Fig. 3, cat. no. 22) has 9% lead. Both the Naukratis and Amathus mirrors (Figs. 3 and 6, cat. nos. 23 and 31) have low levels of lead (1.17% for cat. no. 23 and 0.82% for cat. no. 31) as well as relatively low concentrations of tin (8% and 5% respectively). Reflectivity increases with the tin concentrations, and with such a low tin content, as already observed for contemporary Greek and Etruscan mirrors, these mirrors would have provided a yellowish or rosy reflection. When a silvery image was favoured in Late Hellenistic and Roman times, the mirrors' composition shifted to c. 20% Sn and 10% Pb (Craddock and Buck, 1993; Swaddling, 2001). The bronze bowl (Fig. 3, cat. no. 24) has very little lead (0.02%, impurity) and contains 6.9% tin. It does not reach the high average Sn content observed in plain hemispherical bowls from Palaepaphos-Skales dated from the 11th to the 8th century BC, but lower tin levels for this type of bowl have also been noted (Charalambous et al., 2014; see also high-tin bronze vessels from the Early Iron Age cemetery of Palaepaphos-Plakes analysed by Charalambous and Kassianidou in Karageorghis and Raptou, 2014, 120–122). Straight tin bronze without addition of lead is a composition to be expected with an object shaped by hammering, though many similar high-tin bronze bowls from Cyprus were also cast (Charalambous et al., 2014). Despite its shape being similar to Cypriot bowls, the lead isotope ratios, as we will see below, do not match copper deposits on Cyprus. Finally, the small foundation plaque (Fig. 5, cat. no. 30) is a high tin bronze with 11.4% tin, with lead present only as an impurity (Pb: 0.04%), making this object ideal to discuss the copper origin. 4.3. Weights The composition of the two analysed weights is exceptional. The first weight (Fig. 3, cat. no. 14) contained only 2.59% of lead, against 18% of lead for the second weight (Fig. 3, cat. no. 15). Both were made from unusual alloys, namely Cu-Sn-Sb-Pb and Cu-Sb-Pb, respectively. Since antimony was not recognised as a separate metal in antiquity, it must have been introduced either with the copper or the lead. The socalled fahlores are copper ores with high concentrations of antimony which are usually correlated with high concentrations of arsenic, silver, and bismuth (Tadmor et al., 1995; O'Brien, 2015, 225). Since these elements are low in the two weights it seems more likely that the antimony entered the alloy together with the lead. Indeed, lead ores are often associated with stibnite, sometimes called antimonite, a sulphide mineral with the formula Sb2S3. Pastes of stibnite powder in fat or in other materials have been used since ca. 3000 BCE as eye cosmetics called kohl in Egypt and southwest Asia (Forbes, 1965, 17–19). If a lead ore associated with stibnite is smelted to lead then antimony is also reduced to the metal resulting in a lead‑antimony alloy. Stibnite was occasionally smelted to produce metallic antimony but the metal thus produced was taken as a kind of lead (Forbes, 1950). Small objects of antimony are known from Assur dating to around 2000 BCE and from the Late Bronze Age in the southern Caucasus (Hauptmann and Gambaschidze, 2001). 4.6. Faience The five faience objects included in the present article were originally analysed as a part of a wider study of glazed artefacts from Naukratis and Rhodes with focussed on green and yellow glazed objects (Meek et al., 2016). Initially, it was assumed that for the production of bright green faience objects only copper was added to the glaze mixture to obtain the colour. Lead glazes were known in this period but applied only to clay vessels and building material such as the glazed bricks of the Ishtar Gate in Babylon (Tite and Shortland in Razmjou et al., 2004). However, analyses with proton induced X-ray emission (PIXE) spectrometry, conducted in the C2RMF (Centre de Recherche et de Restauration des Musées de France) (Meek et al., 2016) showed that many of the green-coloured faience objects from Naukratis contained significant concentrations of lead, as well as copper. It is not possible to ascertain whether the lead and copper were added to the faience together, or as two separate raw materials. The yellow glazes are coloured with a combination of lead and antimony, and contain very low copper levels. This means that the lead isotope ratios in the glazes can only be used to discuss the provenance of the lead and not of the copper. Therefore, and because only very small samples could be extracted from the fragile objects, only a small selection of five faience objects was sampled for lead isotope analysis. 4.4. Arrow heads The four arrow heads said to have been discovered in the sanctuary of Apollo are characterized by high levels of lead. With its high content of lead (7.74%), the leaf-shaped arrow-head (Fig. 3, cat. no. 16) was far from being sturdy and was already bent in antiquity. The three other arrows (Fig. 3, cat. nos. 17–19) have even higher lead contents, ranging from 16.5 to 19.7%. Arrow heads usually have higher Sn contents compared to their Pb contents (see for example contemporary arrow heads from Phoenician sites in the Iberian Peninsula containing between c. 7 and 25% Sn against 0.5 to 5% Pb: Giumlia-Mair, 2015). Lead at such high level in the Naukratite specimens could have rendered the arrows ineffective and, with Apollo known as an 328 Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. Table 3 Lead isotope ratios in all analysed samples as determined with MC-ICP-MSa. Cat no. Lab no. Object 1 2b 3 4 5b 6 7 8 9b 10 11b 12 13 14 15 16 17 18 19 20b 21b 22 23b 24b 25 26 27 28 29b 30b 31b 32b 33b 34 MA-154824 MA-154850 MA-154849 MA-154851 MA-154848 MA-154852 MA-154841 MA-154842 MA-154843 MA-154844 MA-154845 MA-154846 MA-154847 MA-154822 MA-154823 MA-154831 MA-154832 MA-154834 MA-154836 MA-154837 MA-154853 MA-154828 MA-154829 MA-154830 MA-154838 MA-154826 MA-154854 MA-154855 MA-154857 MA-154856 MA-154840 MA-154839 MA-154859 MA-154858 Slag Neith statuette Ichneumon figure Eel votive box Mahes statuette Cobra votive box Osiris statuette Isis nursing statuette Wig, figure-fitting Bull statuette Atum statuette Situla Situla (base) Weight Weight Arrow-head Arrow-head Arrow-head Arrow-head Arrow-head Arrow-head Stamp-seal Mirror Bowl Bottle Dowel Lead ore Lead ore Stamp-seal Plaque Mirror Sistrum Worshipper statuette Worshipper statuette a b 208 Pb/206Pb 2.0764 2.0666 2.0773 2.0599 2.0685 2.0686 2.0695 2.0623 2.0655 2.0714 2.0645 2.0864 2.0594 2.0730 2.0592 2.0714 2.0609 2.0629 2.0698 2.0879 2.0753 2.0619 2.0691 2.0674 2.0631 2.0699 1.9956 1.9916 2.0700 2.0313 2.0686 2.0639 2.0868 2.0778 207 Pb/206Pb 0.83693 0.83413 0.83712 0.83154 0.83163 0.83422 0.83419 0.83245 0.83311 0.83314 0.83320 0.84506 0.83160 0.83669 0.83144 0.83423 0.83214 0.83302 0.83387 0.85043 0.83945 0.83181 0.83573 0.83636 0.83308 0.83420 0.79954 0.79752 0.83476 0.82041 0.83428 0.83319 0.84846 0.83803 208 Pb/204Pb 38.895 38.817 38.928 38.827 38.999 38.871 38.871 38.828 38.851 38.987 38.826 38.683 38.812 38.829 38.811 38.920 38.821 38.805 38.902 38.361 38.657 38.854 38.784 38.615 38.823 38.886 38.972 39.000 38.862 38.613 38.843 38.816 38.526 38.899 207 Pb/204Pb 15.678 15.667 15.688 15.674 15.679 15.676 15.668 15.673 15.670 15.681 15.670 15.668 15.672 15.672 15.670 15.675 15.675 15.670 15.672 15.625 15.636 15.674 15.665 15.622 15.677 15.671 15.615 15.617 15.672 15.595 15.666 15.670 15.664 15.689 206 Pb/204Pb Suggested source 18.733 18.783 18.740 18.849 18.854 18.791 18.783 18.827 18.810 18.822 18.807 18.541 18.846 18.731 18.847 18.790 18.837 18.811 18.795 18.373 18.627 18.843 18.745 18.679 18.818 18.786 19.530 19.582 18.774 19.009 18.778 18.807 18.462 18.722 NW Anatolia Faynan/NW Anatolia NW Anatolia Laurion Faynan/NW Anatolia Balya/Thasos Balya/Thasos Laurion Faynan/NW Anatolia Balya/Thasos Faynan/NW Anatolia Nakhlak Laurion Caucasus Caucasus Balya/Thasos Laurion Laurion Balya/Thasos Faynan/Pontids Cyprus Laurion Faynan/NW Anatolia Sinai-Eastern Desert? Laurion Balya/Thasos Egypt Egypt Faynan/NW Anatolia Sinai-Eastern Desert? Faynan/NW Anatolia Faynan/NW Anatolia Faynan/Pontids NW Anatolia Precisions are better than 0.01% for 208Pb/206Pb and 207Pb/206Pb and better than 0.03% for 208Pb/204Pb. Indicates samples with < 4% lead: for those, the suggested source is a copper source while for all others it is a lead source. Fig. 8. Lead isotope ratios in copper-based objects with < 4% lead from this study compared with copper ores from Cyprus (Gale et al., 1997). For numbers see Table 1. copper based on lead isotope ratios (Table 3). Possible sources of copper for Egypt are primarily the deposits of Timna and Faynan in the Arabah as well as the copper deposits in the Sinai Peninsula and the Eastern Desert (Abdel-Motelib et al., 2012). It is also worthwhile to consider Anatolia, Cyprus (Ogden, 2000; Weisgerber, 2006; Yahalom-Mack et al., 2014) and the Arabian Shield including Oman. While strong production evidence for this period is (currently) lacking, it has already been noted as a source of copper production during the Late Bronze Age 5. Results: lead isotope analysis 5.1. Metal objects and ores 5.1.1. Copper origins As discussed above, the origin of the copper can only be considered when the lead concentration is low. There are 13 samples with less than about 4% lead and these can be assessed in view of the provenance of 329 Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. Fig. 9. Lead isotope ratios in copper-based objects with < 4% lead from this study compared with copper ores from the Sinai Peninsula, Eastern Desert as well as Timna and Faynan in the Arabah valley (Gale et al., 1990; Hauptmann et al., 1992; Abdel-Motelib et al., 2012). For numbers see Table 1. and later (Roman, Islamic) periods, and may have played a role in Egyptian copper supply during the New Kingdom (Weeks et al., 2009; Begemann et al., 2010; Liu et al., 2015; Rademakers et al., 2017). Considering the mines of Timna and Faynan, recent research has demonstrated the continuity and even intensification of copper mining in this region during the Early Iron Age (Weisgerber, 2006; Ben-Yosef et al., 2012; Levy et al., 2012) and recently it turned out that the copper of eleven tripod cauldrons from the Zeus sanctuary in Olympia, dating between ca. 950 and 750 BCE, was produced in Faynan in the Wadi Arabah (Kiderlen et al., 2016). Furthermore, the copper production on Cyprus increased during the Late Period (Kassianidou, 2013; Masson, 2015a) and the island should certainly be considered as possible source region for the copper in Egypt. However, from Fig. 8 it is clear that the lead isotope ratios of only one sample (the arrow-head cat. no. 21) matches with the welldefined (Stos-Gale et al., 1997) isotope range of copper ores on Cyprus. It is the only object that consists of unalloyed copper and the trace element pattern, especially the arsenic concentration, is consistent with Cyprus and not with Faynan, if we use the average composition of oxhide ingots (Hauptmann et al., 2002) as reference. There is one more object plotting together with copper ores from the Larnaka axis (Stos-Gale et al., 1997) on Cyprus (the Mahes statuette cat. no. 5). However, in the plot with 204Pb (Fig. 8b), the trace element pattern and the relatively high lead content (2.16%) are strong indications that the copper does not derive from Cyprus. In theory, one could also find matching lead isotope ratios in copper ores from southeastern Spain in the Almeria province (Stos-Gale et al., 1995). Long before the Roman intense exploitation of the large deposits of Rio Tinto in the Huelva province, the Phoenicians were already exploiting mines in the Iberian Peninsula in the early 1st millennium BC, as demonstrated by measurements of lead pollution in deep Greenland ice and atmospheric modelling (McConnell et al., 2018). Our knowledge is, however, limited as to which mines were exploited at the time and to what extent the extracted ores supplied a regional or a more international market. Geographically the most likely candidates for the provision of copper are the copper deposits in the Arabah valley, the Sinai Peninsula, and the Eastern Desert. The lead isotope ratios of these deposits are compared with the copper-based objects from Naukratis, Tell Dafana and Cyprus in Fig. 9. It seems that Timna can be excluded as source but the three other regions do exhibit some overlap with the objects. The potential for copper production in the Sinai and Eastern Desert usually cannot compare with other copper districts such as Cyprus or Faynan with the exception of Bir Nasib, located in the southwestern Sinai (e. g. Hauptmann, 2007). Although Bir Nasib is poorly investigated and often underestimated as a major copper supplier (Stos, 2009), extensive smelting activities took place at the site at least during the LBA/EIA. Lead concentration in the Sinai and Eastern Desert copper ores are characteristically very low (Abdel-Motelib et al., 2012), rarely exceeding 0.1%. Only the bowl from Naukratis (cat. no. 24), containing 0.02% Pb, matches the chemical and isotopic values of Bir Nasib deposits. The copper from the Tell Dafana plaque (cat. no. 30), which also contains little lead (0,04%), might as well originate from the Sinai or Eastern Desert deposits, but the paucity of data from these regions makes it uncertain. Similarly, although there is seemingly little overlap between the copper ore from Faynan and the selected copper-based objects, one has to bear in mind that the whole range of lead isotope ratios of Faynan is not yet so well explored as e.g. on Cyprus, and, it seems that Faynan provides one of the best overall compatibilities with the objects. It remains to check if the copper could derive from the Arabian shield or Anatolia. Nothing is known about possible prehistoric exploitation of several massive sulphide deposits in Saudi Arabia, for which lead isotope data have been published (Doe and Rohrbough, 1977; Stacey et al., 1980; Bokhari and Kramers, 1982). Their 207Pb/204Pb ratios are generally much lower (around 15.5 or lower) than the selected copper-based objects that range between 15.6. and 15.7, so that they need not be further regarded. Copper ores from Oman partly overlap in their lead isotope characteristics (not shown) with our selection of copper-based objects but they are, like the copper ores from Sinai, very low in lead concentrations with the exception of Nujum where there is also an ancient lead mine which was also exploited for copper ores, at least in the eastern parts (Prange, 2001; Begemann et al., 2010). However, isotopically these ores do not fit the selected copper-based objects (Fig. 10). Also shown in Fig. 10 are the lead isotope ratios of copper ores from Anatolia. It has been observed that base metal deposits in the three major tectonic units in Anatolia can be differentiated by their lead isotope ratios albeit with overlaps (Pernicka et al., 1990). In principle, all selected copper-based objects are consistent with these copper deposits, with most samples matching the isotopic values of the northwestern Anatolian ores. Evidence for prehistoric mining is available in all three regions, e.g. Ergani Maden in the Taurids, Küre in the Pontids and Serçeörenköy in northwest Anatolia (Pernicka et al., 1984; Wagner et al., 1989; Wagner and Öztunalı, 2000). These regions are not as rich in copper as the major copper districts of Cyprus or Faynan. Nonetheless, Anatolian copper could have been significant in Egypt at a time when the Greeks settled in Western Anatolia played such an active role in trade, as is particularly well documented historically and archaeologically at Naukratis in the Late Period (see below). 330 Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. Fig. 10. Lead isotope ratios in copper-based objects with < 4% lead from this study compared with copper ores from Anatolia (Seeliger et al., 1985; Wagner et al., 1986; Begemann et al., 2003) and from Oman (Begemann et al., 2010). notably in Turkey and/or the Aegean and Tunisia. The lead in the group of objects plotting in the ellipse most likely derives from Laurion in Attica (Fig. 12). One group of three objects (the slag sample cat. no. 1 and the ichneumon mummy-case cat. no. 3 from Naukratis, and the statuette of worshipper from Idalion cat. no. 34) plot together, having their best matches of ore sources in northwest Anatolia, notably at the huge deposit of Gümüşköy (Pernicka et al., 1984), which was probably exploited since prehistoric times. Another group of five leaded tinbronzes from Naukratis (two arrow-heads cat. nos. 16 and 19, the Osiris statuette cat. no. 7, the bull statuette cat. no. 10 and the cobra votive box cat. no. 6) plots very closely together with the lead dowel cat. no. 26. The best match of lead ore sources are Balya in northwestern Anatolia (Pernicka et al., 1984) and the island of Thasos in the northern Aegean (Vavelidis et al., 1988), both known to have been in use in the Greek period. The lead in the situla from Naukratis (cat. no. 12) seems to derive from yet another source and it may be surprising that there are many matches with lead ores and lead slag from Nakhlak in Iran as well as with litharge (lead oxide from cupellation for silver production) in Arisman south of Kashan in Iran (Pernicka et al., 2011). Nakhlak has been an important production site of lead and silver since the fourth millennium BC and is still in operation today. According to the new archaeological investigations, Naukratis was a thriving port during the Persian domination of Egypt. Imports of metals (or metal objects) from 5.1.2. Lead origins Concerning the question of the origin of the lead in the lead-rich objects it is obvious from Fig. 11 that the two galena samples most likely derive from Egypt due to the highly radiogenic lead, which is frequent for Egyptian lead deposits. Activities in lead mines are reported at the beginning of the 26th dynasty at a lead mine at Gebel el Rosas along the Red Sea coast. Inscriptions found there record that Montuemhat, fourth prophet of Amun and mayor of Thebes, sent an expedition to the lead mines during year 14 or 15 of the reign of Psamtik I, in 650 or 649 BCE (Vikentiev, 1957; Masson, 2015a, 80). The 208Pb/206Pb and 207Pb/206Pb values for lead ores from Gebel el Rosas are, however, too high for the lead ore samples uncovered in 26th dynasty contexts at Tell Dafana. The large mining site of Gebel Zeit in the Eastern Desert opposite Sinai, which was already exploited in the Late Bronze Age, seems a better candidate; but there are not enough ore body LIA for Egyptian ores to point with certainty to a specific mine (Shortland, 2006, 2009). The main sources of lead, however, were not in Egypt (Fig. 12), corroborating the results of the analyses by Fleming (1982) and by Schwab and Willer (2016) (Fig. 13). The two lead objects from Naukratis were produced from lead sourced outside Egypt, at Laurion in Attica (bottle cat. no. 25) and another location (lead dowel cat. no. 26, see below). Also the lead in six of the 14 leaded copper alloyed finds can be attributed to Laurion and the remaining to other lead deposits Fig. 11. Lead isotope ratios in the lead-rich objects from Naukratis (red crosses) and Cyprus (red cross in an open circle) compared with kohl and lead ores from Egypt (Stos-Gale and Gale, 1981; Shortland, 2006) and from Laurion in Attica, Greece (Stos-Gale et al., 1996). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) 331 Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. Fig. 12. Lead isotope ratios in the lead-rich samples from Naukratis (red crosses) and Cyprus (red cross in an open circle) compared with Anatolian lead ores (Doe and Rohrbough, 1977; Gale, 1979; Seeliger et al., 1985; Wagner et al., 1985; Wagner et al., 1986; Yener et al., 1991; Sayre et al., 1992). The ellipse in Fig. 12a marks the spread of lead isotope ratios in the lead‑silver deposits of Laurion in Attica, Greece. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) Iran to Naukratis during that period are therefore not impossible. To find scientific support for such a hypothesis is nonetheless significant. 6. Discussion 6.1. Cyprus Among the 13 metal finds for which a likely copper provenance could be established, only the oblanceolate tanged arrow-head (cat. no. 21) was produced with Cypriot copper. This Egyptian type of arrowhead was most likely produced in Egypt using imported Cypriot copper during the Saite period. The limited evidence for finds made with Cypriot copper at Naukratis (either imported or made locally) is unexpected, especially since Cypriot imports are otherwise well attested at Naukratis. Over 250 terracotta and stone votive figures and figurines imported from Cyprus have been found in the Greek sanctuaries of Naukratis, the bulk of them dating back to the earliest history of the site (Nick, 2006; Höckmann, 2007; Thomas, 2015b). Cypriot basket-handle amphorae and mortaria are regularly noticed in Late Period contexts, both during the early and most recent excavations conducted at the site (Johnston, 1982; Villing, 2006, 2015). In addition to the obvious trade links between Naukratis and Cyprus, copper mining in the Cypro-Archaic and Cypro-Classical periods is now very well documented thanks to a series of recent surveys and studies (with the earliest C-14 dates for 5.2. Faience finds Unlike for LBA faience (Shortland, 2006), local lead ores seem not to have been used in mass-produced amulets during the Late Period, at least as far as we can tell from the case of Naukratis. The lead isotope ratios in all faience objects analysed are consistent with Laurion (Table 4 and Fig. 14). For the green glazed objects (cat nos. 35, 37 and 39) it is possible that there may be a contribution of lead added as part of the copper colouring material. However, the relatively high Pb:Cu ratio for these items (Meek et al., 2016; see also the composition table in annex, Table 5) suggests that the bulk of the lead is from a separate addition. Two samples plot just outside the scatter field of lead ores from Laurion, which could possibly be explained by some contamination from another raw material used in their production. The closest LI field remains nonetheless Laurion for these objects. Fig. 13. Lead isotope ratios in the lead-rich samples from Naukratis (red crosses) and Cyprus (red cross in an open circle) compared with other lead-rich objects from Egypt and the Sudan analysed by Fleming (1982) and by Schwab and Willer (2016). The ellipse in Fig. 13a marks the spread of lead isotope ratios in the lead‑silver deposits of Laurion in Attica, Greece. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) 332 Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. Table 4 Lead isotope ratios in the glazes of faience objects from Naukratis as determined with MC-ICP-MS. Precisions are better than 0.03% for 208Pb/206Pb and 207Pb/206Pb and better than 0.05% for 208Pb/204Pb. Cat. no. Lab no. Object 208 Pb/206Pb 207 Pb/206Pb 208 35 36 37 38 39 MA-155408 MA-155409 MA-155410 MA-155411 MA-155412 New Year's flask Scarab Scarab Scarab Scarab 2.0656 2.0590 2.0604 2.0659 2.0595 0.83315 0.83147 0.83209 0.83308 0.83150 38.854 38.790 38.814 38.870 38.813 mining assigned to the 10th and 9th centuries, so in the Cypro-Geometric period: Kassianidou, 2012, 2013). The copper-rich central region of Cyprus to the North-West of Tamassos and in the region of the Troodos mountains, seems to have been particularly active (see for example Given and Knapp, 2003; Given et al., 2007; Ben-Yosef et al., 2011). Pb/204Pb 208 Pb/204Pb 15.671 15.664 15.675 15.674 15.670 208 Pb/204Pb 18.810 18.839 18.838 18.815 18.846 but also Levantine—from Faynan and Timna—and unidentified sources: Bartelheim et al., 2008, 2011). The exploitation of copper ores from the Sinai was extensive throughout the BA, particularly during the Old and Middle Kingdoms, but also during the New Kingdom (Rademakers et al., 2017). Post-BA exploitation of copper in the Sinai, as well as in Eastern Desert, has been spotted at a few mines (Abdel-Motelib et al., 2012). In the southwestern Sinai, the massive smelting site of Bir Nasib was still active in the EIA while Wadi Homr showed signs of exploitation in the EIA and again later in the Roman or medieval times. Ptolemaic and/or Roman exploitation of copper ores is also attested at Gebel Dara in the Northern Eastern Desert and Wadi Hamama in the Central Eastern Desert. The apparent absence of Late Period activities on the copper deposits of the Sinai and Eastern Desert could be explained by the poor investigation of these regions. There is some overlap between the lead isotope ratios of the copper-based objects from this study with those from New Kingdom site Qantir (Fig. 15), but what Rademakers et al. (2017) described as ‘Egyptian’ copper that is also found in Amarna is isotopically different from Naukratis, with slightly higher 206Pb/204Pb ratios. If our identification is correct for cat. nos. 24 and 30, then this would be the first verification of the use of Sinai/Eastern Desert copper reflected in 26th dynasty metal artefacts. In any case, the activities in the lead mines of the Eastern Desert, already documented by the aforementioned mid-7th century BC inscription, seem further validated: the two galena samples found in 26th dynasty contexts at Tell Dafana (cat. nos. 27–28) are identified as Egyptian lead ores. The use of Wadi Arabah copper is now well attested in the EIA. Faynan copper was identified in the loaf-shaped ingots recovered from a LBA/EIA wreck off the Carmel coast in Israel (Yahalom-Mack et al., 2014). Wadi Arabah copper was also recently recognised on various copper alloy objects from an EIA sanctuary in the Phoenician city of Sidon (Lebanon), in activity between the 11th to the 8th century BC 6.2. Faynan and the Sinai/Eastern Desert The copper ores from the Wadi Arabah, and particularly Faynan, represent a rather good match for most metal finds for which copper provenance could be established, while the finds with a Pb content lower than 0.1% could be congruent with the Sinai Peninsula or the Eastern Desert copper deposits. The finds belong to various categories, and were found in Naukratis, Tell Dafana and Cyprus. They include objects of definite Egyptian manufacture (cat. nos. 2, 5, 9, 11, 29 and 30), as well as objects that could have been produced either in Egypt or imported from the Eastern Mediterranean (cat. nos. 20, 23 and 31). The hemispherical bowl (cat. no. 24) was thought to be a Cypriot import, but the non-Cypriot origin of its copper makes this maybe less probable. Likewise, Egyptian and Egyptianizing objects discovered in Cyprus do not contain Cypriot copper. The Hathoric sistrum (cat. no. 32) is thus almost certainly an import from Egypt. The Cypro-Archaic statuette of a worshipper (cat. no. 33) is a type of Egyptianizing statuette does not fit the types of bronze votives produced in Egypt at the time, but rather resembles Phoenician metalwork and is well-attested in Cyprus and elsewhere in the Mediterranean (Kiely, 2014; Jiménez Ávila, 2015; Bernardini and Botto, 2015). This is not the first time that bronze objects discovered in Cyprus are found to contain metal that did not derive from Cyprus (Stos-Gale, 2000, 66–68; see also LBA hoard of metal vessels and other objects from Kalburnu on the Karpas peninsula with objects made from a wide array of sources of copper, not only Cypriot Fig. 14. Lead isotope ratios in faience glazes from Naukratis compared with lead ores from Laurion in Attica, Greece (Stos-Gale et al., 1996) at high resolution. No Egyptian lead ores plot in this small section of the general lead isotope diagram. Note that experimental uncertainties are larger than the data points in this diagram. For numbers see Table 1. Experimental uncertainties are slightly larger than the symbols. 333 Journal of Archaeological Science: Reports 21 (2018) 318–339 0.01 0.02 0.02 0.01 0.01 0.02 0.03 0.03 0.02 0.04 0.00 0.01 0.01 0.01 0.01 0.16 0.30 0.00 0.01 0.10 0.56 10.35 0.24 0.72 0.51 0.01 0.00 0.02 0.00 0.00 1.90 7.91 1.47 1.25 1.89 (Vaelske et al., forthcoming). Other recent analyses carried out on a large corpus of tripods from Greek sanctuaries spanning the period between 1100 and 700 BCE indicated that Wadi Arabah copper dominated the markets of Greek mainland till about 750 BCE, with a marked shift in the Late Geometric (760–700 BCE) towards the import of Fahlore-type copper, possibly from the Alps (Kiderlen and Bode, forthcoming). All our selected finds are more recent in date and could demonstrate the continuity of the use of Faynan copper in the Archaic and Classical periods in other regions of the Eastern Mediterranean. However, as we saw, Anatolian ores offer a good match for the copper-based objects of our study too. 6.3. Anatolian and North Aegean regions All selected copper-based objects we have just discussed, but also many of the lead-rich analysed objects, can be found consistent with Anatolian deposits. A first group of three objects – one slag and one Egyptian votive bronze from Naukratis (cat. nos. 1 and 3) as well as an Egyptianising statuette from Idalion (cat. no. 34) – has affinities with sources in northwestern Anatolia, possibly at Gümüşköy. The lead of six more objects from Naukratis is consistent with sources in Thasos or Balya, in the north Aegean and northwest Anatolia, respectively. This group includes objects likely produced in Egypt, possibly at Naukratis itself, such as the lead dowel (cat. no. 26), the three Egyptian votive bronzes (cat. nos. 6, 7 and 10) and an Egyptian-style arrow-head (cat. no. 16). Another arrow-head (cat. no. 19), characteristic of Anatolia and the Ionian cities of western Turkey, was probably produced there using local sources of lead and imported to Naukratis. However, this object and one of the Egyptian votive bronzes (cat. no. 10) have such a low tin concentration that these leaded tin-bronzes could have been produced from various metal scraps, making an LI attribution more tenuous for these two finds. LIA performed on Qubbet el-Hawa objects have established Thasos as the most probable origin for the lead of some Egyptian bronzes locally produced in the 6th–5th century BC (Schwab and Willer, 2016). Although Schwab and Willer admitted that Anatolian lead ore might be an alternative source, they were not aware of any Anatolian lead exports during the 6th and 5th centuries BC. There is abundant evidence, however, for lead‑silver mining and production between the Archaic and Hellenistic periods in the north Aegean and northwestern Anatolia (Pernicka et al., 1984; Kassianidou, 2012, 245–246). Late Period treasures in Egypt have produced a large amount of silver coins from the Thraco-Macedonian region (Möller, 2000, 209; Masson, 2016). That metals trade played a role already in earlier periods is suggested by the analysis of a 7th century BC silver hoard discovered in the Philistine city of Ekron, which identified several sources, among them Laurion, Siphnos, Chalkidice/Troad, Thasos and Chios (Stos-Gale, 2001, 61–62). Regarding trade between the Northern Aegean and Egypt, there is increasing evidence to suggest that East Greek traders carried silver acquired in Thrace or Macedonia south to the Levant and Egypt, to be exchanged for grain or other goods (Roebuck, 1950; Moreno, 2007, 309–315; van der Wilt, 2010). Archaeology confirms that by the 7th century BC the Thracian and northern Greek region was closely tied into Greek trade networks (Tiverios, 2008; Vacek, 2017). It was the great East trading and seafaring cities and islands located on eastern littoral of the Aegean – Miletos, Samos, Teos, Phokaia, Chios and others – that were the driving force in these networks, the same cities that, according to 5th century BC Greek historian Herodotus (Histories 2.178–179) were the ‘founding’ cities of Naukratis (Möller, 2000, 75–88). This is amply confirmed by archaeological evidence from the site, which includes plentiful finds of pottery brought to Egypt from the Greek cities of western Anatolia, including Aeolis and the Troad, in the hinterland of which lay the site of Balya (Villing, 2013b; Villing et al., 2013–19). Hence, the use of lead imported from northwest Anatolia and northern Aegean, as well as maybe the use of Anatolian copper, should not be surprising in Late Period Egypt. Co, Ni, and Rb were sought but were below 0,01% an all samples. EA58327 EA23617 EA66454 EA66477 EA66486 BM BM BM BM BM 35 36 37 38 39 New Year's flask Scarab Scarab Scarab Scarab Green Yellow Green Yellow Green 1.91 0.48 1.19 1.41 1.75 1.45 0.24 0.73 1.20 0.75 4.93 0.90 2.36 5.92 3.26 71.42 71.35 79.35 79.98 75.37 0.81 0.43 0.54 1.79 0.50 5.79 2.12 5.18 1.27 4.72 2.55 0.97 1.24 0.96 1.82 0.56 0.35 0.52 1.65 1.03 5.30 3.61 5.61 2.46 5.89 0.29 0.21 0.07 0.11 0.09 0.01 0.00 0.00 0.01 0.01 0.10 0.01 0.01 0.02 0.03 1.47 0.39 0.53 0.88 0.70 0.74 0.30 0.86 0.32 1.49 ZrO2 SrO Cl SO3 P2O5 SiO2 Al2O3 MgO Na2O Glaze colour Object type Museum number Cat no. Table 5 PIXE-PIGE results for glaze areas analysed from faience objects from Naukratis. Results are normalised to 100%. K2O CaO TiO2 V2O3 MnO Fe2O3 CuO ZnO SnO2 Sb2O5 BaO PbO A. Masson-Berghoff et al. 334 Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. Fig. 15. Lead isotope ratios in copper-based objects with < 4% lead from this study compared with copper-based objects from Qantir (Rademakers et al., 2017). 6.4. Laurion 6.5. Caucasus mountains Lead from Laurion was identified in a variety of objects found at Naukratis. This group includes two heavy barbed and tanged arrowheads produced locally or imported from Greece to be deposited in a Greek sanctuary at Naukratis (cat. nos. 17–18). It is uncertain if the stamp seal in the shape of a plumed cartouche with an Aramaic inscription (cat. no. 22) and the lead bottle (cat. no. 25) were imported or produced locally. However, Laurion lead was also recognised for Egyptian votive bronzes (cat. nos. 4, 8 and 13), all produced in Egypt, possibly at Naukratis itself. Furthermore, it appears that faience scarabs and New Year's flasks manufactured at Naukratis during the 26th dynasty used Laurion lead in their production. Our results compare well with other recent research that has highlighted the importance of Laurion lead. LIA conducted on 59 magical artefacts, principally made of lead and dated to the 4th century BC, revealed that an overwhelming majority used lead from Laurion mines (Vogl et al., 2016). Most of these objects come from the Greek world, especially from Attica. Eight analysed samples from six curse tablets found or bought in Egypt showed in six cases the use of Laurion lead. The authors concluded that they were imported into Egypt from Attica (Vogl et al., 2016, 15–16), as also the tablets' typology and contents fit that of the Attic curse tablets. We should nonetheless not entirely discard the possibility that they could have been produced in Egypt itself using lead imported from Laurion. The discovery of loaf-shaped stamped lead ingots at ThonisHeracleion, the sister harbour town of Naukratis on Egypt's Mediterranean coast, corroborates the wide circulation of lead from Laurion and its import in Egypt (van der Wilt, 2010). Their shape and stamps are similar to lead ingots found in the Porticello wreck and LIA suggested the mines of Laurion as the source for the Porticello wreck's ingots (Eiseman and Ridgeway, 1987, 57, 107). Laurion lead seems indeed to have been commonly used in the production of metal objects in Late Period Egypt, as indicated also by LIA conducted on several copper-based finds from more southern Egyptian and Nubian sites (Fleming, 1982; Schwab and Willer, 2016). LIA carried out on a Phoenician lead cup belonging to the Persian period and discovered on the coast of Byblos further identified Laurion as the best match for the source of lead (Gubel and Cauet, 1987). With our results, however, it is the first time that the use of imported lead is recognised in the production of Egyptian faience amulets and vessels. Obviously, the case of Naukratis could be exceptional: lead mines in the Eastern Desert were exploited during the 26th dynasty and it is possible that local lead ores were used in other faience workshops without access to imported lead. The unusual lead‑antimony alloy recognised for the two analysed weights (cat. nos. 14–15) led us to suggest as a possible origin the southern Caucasus, to the north-east of the Black Sea. Their dome shape is particularly related to Egyptian types of weight, but it is also well attested in the Near East. Therefore, it remains uncertain if these weights were produced in Egypt or imported from elsewhere. Metals and metallurgy in the Caucasus were crucial elements in exchange and cultural interactions since the fifth millennium BC, particularly since around 3000 BCE (Courcier, 2014). Caucasian antimony metal was widely exploited during antiquity since the Bronze Age (O'Brien, 2015). Large-scale antimony mining is particularly well-documented in the LBA and Early Iron Age, especially in the Gornaya Racha region of Transcaucasia and its major antimony mine Zopkhito (Chernykh, 1992, 60, 113 and 276; Pike, 2002). The use of Caucasian antimony has also been suggested by Shortland in his study of LBA glass produced at Malqata and Amarna in Egypt (Shortland, 2002). According to the available data at present, while Iran and Anatolia are both possible origins for stibnite, only Caucasian mines were already exploited in antiquity (Shortland, 2012, 112–113). The wealth of natural resources available in the Pontic region, including metals, has often been suggested as the major reason for why numerous Greek colonies were founded on the shores of the Black Sea and beyond in the Archaic period, although this is a hotly debated topic (Tsetskhladze, 1995, 1998; Treister, 1998; Greaves, 2007). Many of these Greek settlements were established by Miletos (Greaves, 2007) and other Ionian cities. Relationships between Greek settlers and local population seem largely to have been peaceful according to latest research (e.g. Tsetskhladze, 2016). Some of the Greek settlements were located close to metalliferous mineral reserves, such as the Milesian colony of Apollonia Pontica in Bulgaria next to the copper mines of Medni Rid (Baralis et al., 2015). Recent archaeological investigations in the Archaic mines of Medni Rid only uncovered Archaic Greek material, while the 6th century BC levels within the city yielded substantial amounts of slags. This suggests that the inhabitants of Apollonia Pontica exploited these mines (Baralis et al., 2016). Since the mines were located in a sector administered by Thracian communities, there must have been some agreements with Thracian populations, at least for the 6th century BC. We have already stressed that Eastern Greek cities were involved in the trade at Naukratis and that includes Miletos, one of the ‘founding cities’ of Naukratis according to Herodotus. They could very well have been the agents of these imports of northwestern Anatolian and Caucasian metals, directly or indirectly through their colonies. This 335 Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. trade link is further supported by the presence of Naukratite scarabs in the Black Sea region (Gorton, 1996), revealing once again how the harbour-town of Naukratis was crucial in the transit and distribution of goods between Egypt and the wider Mediterranean world. that lead ores from the Eastern Desert were not used in the production of metal artefacts and perhaps neither in that of faience objects (although the later statement is only based on a few faience finds from Naukratis). Yet, the ore samples from Tell Dafana confirm the exploitation of galena from the Eastern Desert during the 26th dyansty, but galena could have had other applications (possibly kohl). Instead, lead from Laurion in Attica, the North Aegean and northwestern Anatolia was detected, and even perhaps from the southern Caucasus and Iran. The diversity of the lead sources is indicative of a complex trading network for metal in which Greeks likely played a major role. The results show that Naukratis, alongside its sister harbour ThonisHeracleion, was the conduit through which imported goods entered Egypt and from there were redistributed, as is indicated by the presence of Laurion and Western Anatolian lead in 6th–5th century leaded bronzes from Qubbet el-Hawa in the south of Egypt. But they also open up new, wider questions. Further research is needed into patterns of metal ore exploitation and trade in the first millennium BC, including the role of Egyptian, Anatolian and Eurasian resources. To what extent were Greek and/or Phoenician traders involved in the trade of copper and lead to Egypt more generally? Would similar profile of metal import be encountered on other towns and regions of Egypt? What was the role of Anatolian or Faynan and Sinai/Eastern Desert copper in the wider Mediterranean world during the Archaic and Classical periods? The results obtained on finds from Tell Dafana and Cyprus seems to indicate that patterns observed at Naukratis extend to wider regions, but the pool of comparative samples is far too limited. Understanding the intense, complex trade of the Archaic and Classical periods in the Mediterranean world requires a large-scale research, the necessity and potential of which this pilot project has demonstrated. 6.6. Iran The lead source of one of the seven situlae found at Naukratis (cat. no. 12) has been tentatively identified as the site of Nakhlak in Iran. Situlae, both models like this one and full-sized versions, were standard mass-produced votives of Late Period and Ptolemaic Egypt. Although the Naukratis situlae have numerous parallels not only in Egypt, but also in Nubian, Cypriot, Greek and Near-Eastern contexts (Bell, 2011, 406–416), the situla cat. no. 12 was most likely produced locally: one of the figures decorating its walls represents the god Min or Amun-Min, an ithyphallic deity revered at Naukratis (Masson, 2015b). With a possible Iranian origin for its lead, it would be tempting to date the situla to the 27th dynasty (525–404 BCE) or the 31st dynasty (343–332 BCE), when Egypt was under the dominion of the Achaemenid Persian Empire. The Achaemenid period (550–330 BCE) saw an increase in the exploitation of metal deposits in Iran, notably for copper, gold, silver and lead‑zinc (Ghorbani, 2013, 67–68). Could this period have seen a shift in the sourcing of metals for Egypt? However, the low content of tin (0.95%) probably signals that the situla was produced from recycled metal. It seems doubtful therefore that lead was directly imported from Iran. If tin was added as cassiterite, an alternative interpretation could be the use of a low-quality concentrate which would introduce more iron than tin. 7. Conclusions Acknowledgements The initial impetus to conduct this study was the sudden increase in the course of the 26th dynasty in the production of objects in copper alloys, and also in glazed composition (though to a far lesser degree with copper and lead only needed for the colourful glazing), suggested a rather stable and regular supply of raw material rather than only reliant on the recycling of bronze scraps. As a centre of cultural, technical and commercial exchange between Egypt and the Mediterranean world, Naukratis represented an ideal starting point for this investigation of the supply of metal into Egypt. The new chemical and lead isotope data on a wide range of metal and faience objects from Naukratis, and several finds from Tell Dafana and Cyprus, confirms to some extent our initial hypothesis of metal trade and indicates that this supply came from a variety of sources, which we were able to pinpoint with varying degrees of certainty. Future data on ore deposits may firm up or modify some of our conclusions, and given the dearth of LIA performed on Egyptian objects, it is still too early to draw up a complete map of metal trade in Egypt during the Late Period. Yet the data presented here gives an important first insight into key trends, including some unexpected patterns. For copper, the Wadi Arabah or possibly the northwestern Anatolian regions, as well as Cyprus and maybe the Sinai/Eastern Desert were identified as likely sources. Further analyses would be needed to firm up the identification of copper sources used in Egypt and to conclusively exclude the importance of recycling in this period. The identification of non-Cypriot copper for objects found in Cyprus suggests their Egyptian and/or Phoenician manufacture. Despite the results of the Cypriot items being disconcerting at first glance, they are not that surprising. Dynamic copper and bronze producing areas may well have made use of different sources to ensure supply and/or imported finished products as models (or as prestige goods) which were copied locally. We also need to emphasise that the types we have selected for analysis are ambiguous. Although it would be too exaggerated to attribute a lot of bronzes found in Cyprus to ‘Phoenicians’ (as recently done in Vonhoff, 2015), an on-going interplay between Cyprus, the Levant and Egypt in this period could have developed due to trade in both the metal, the technology and the goods. As regards lead, it appears clearly This study has been made possible by a grant of the Gerda Henkel Stiftung (AZ07lVl15), which we gratefully acknowledge. This paper has greatly benefited from fruitful comments and discussions with specialist of Cypriot material culture Thomas Kiely, and colleagues from the Naukratis project Alexandra Villing and Ross Thomas (Greece and Rome department, British Museum). We would like to thank Paul Roberts and Liam McNamara (Ashmolean Museum), as well as Alice Stevenson and Pia Edqvist (Petrie Museum) for allowing and facilitating the sampling of objects from their museums. We are indebted to Sigrid Klaus and Bernd Höppner (CurtEngelhorn-Zentrum Archäometrie, Mannheim) for their support in the laboratory work. We are grateful to Sylwia Janik and Claire Messenger (Ancient Egypt and Sudan department, British Museum) and Elka Duberow (Curt-Engelhorn-Zentrum Archäometrie, Mannheim) for helping out on the administrative aspects of the project. At last, the authors would like to acknowledge the two anonymous reviewers whose comments have greatly helped improving this manuscript; any remaining shortcomings are the authors' responsibility. Funding This work was supported by a research grant from the Gerda Henkel Foundation (AZ07lVl15). References Abdel-Motelib, A., Bode, M., Hartmann, R., Hartung, U., Hauptmann, A., Pfeiffer, K., 2012. Archaeometallurgical expeditions to the Sinai Peninsula and the Eastern Desert of Egypt (2006, 2008). Metalla (Bochum) 19, 3–59. Agut-Labordère, D., 2012. Le statut égyptien de Naucratis. In: Dieudonné, V., Feyel, C., Fournier, J., Graslin, L., Kirbilher, F., Vottéro, G. (Eds.), Entités locales et pouvoir central: La cité dominée dans l'Orient hellénistique, Nancy, pp. 353–373. Aja, A.J., 2011. Metal weapons and tools. In: Stager, L.E., Master, D.M., Schloen, J.D. (Eds.), Ashkelon 3: The 7th Century BC, Indiana, pp. 503–545. Auenmüller, J., 2014. Die Ergebnisse der bildgebenden Verfahren. In: Fitzenreiter, M., 336 Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. Doe, B.R., Rohrbough, R., 1977. Lead isotope data bank: 3459 samples cited. US Geol. Surv. Open-file Rep. 77–418. Eiseman, C.J., Ridgeway, B.S., 1987. The Porticello Shipwreck: A Mediterranean Merchant Vessel of 415–385 B.C. Nautical Archaeology Series 2, Austin. Erdmann, E., 1973. Die sogenannten Marathonpfeilspitzen in Karlsruhe. Archäologischer Anzeiger 88, 30–58. Faegersten, F., 2003. The Egyptianising Male Limestone Statuary from Cyprus. A Study of Cross-Cultural Eastern Mediterranean Votive Type, Lund. Figueiredo, E., Silva, R.J.C., Senna-Martinez, J.C., Fátima Araújo, M., Braz, F.M., Inêz Vaz, J., 2010. Smelting and recycling evidences from the Late Bronze Age habitat site of Baiões (Visey, Portugal). J. Archaeol. Sci. 37, 1632–1634. Fitzenreiter, M., Willer, F., Auenmüller, J., 2016. Materialien einer Gusswerkstatt von der Qubbet el-Hawa. Bonner Aegyptiaca, Berlin. Fleming, S.J., 1982. Lead isotope analyses of Late Period Egyptian bronzes. MASCA Journal 2 (2), 65–69. Fleming, S.J., Crowfoot-Payne, J., 1979. PIXE analyses of some Egyptian bronzes of the Late Period. MASCA Journal 1 (2), 46–47. Forbes, R.J., 1950. Metallurgy in Antiquity: A Notebook for Archaeologists and Technologists, Leiden. Forbes, R.J., 1965. Studies in Ancient Technology III (Second Edition), Leiden. Gale, N.H., 1979. Lead Isotopes and Archaic Silver Coins. Archeo-Physika 10, Bonnpp. 194–208. Gale, N.H., Bachmann, H.G., Rothenberg, B., Stos-Gale, Z.A., 1990. The adventitious production of iron in the smelting of copper in Timna, W. Arabah. In: Rothenberg, B. (Ed.), The Ancient Metallurgy of Copper. Researches in the Arabah 1959–1984. Vol. 1. IAMS Monograph, London, pp. 182–191. Gale, N.H., Stos-Gale, Z., Houghton, J., Speakman, R., 1997. Lead isotope data from the Isotrace laboratory, Oxford: Archaeometry data base 4, ores from Cyprus. Archaeometry 39, 237–246. Gale, N.H., Stos-Gale, Z.A., 1982. Bronze Age copper sources in the Mediterranean. Science 216 (15–17), 11–19. Gale, N.H., Wagner, G.A., Gentner, W., 1980. Mineralogical and Geographical Silver Sources of Archaic Greek Coinage. In: Metcalf, D.M., Oddy, W.A. (Eds.), Metallurgy Numismatics. Roy. Num. Soc. Special Publication No.13, London, vol. 1. pp. 3–50. Gale, N.H., Stos-Gale, Z.A., 2000. Copper Oxhide Ingots and the Mediterranean Metals Trade: A New Perspective. In: Ioannides, G.K., Hadjistyllis, S.A. (Eds.), Acts of the Third International Congress of Cypriot Studies. Society of Cypriot Studies, Nicosia, pp. 327–344. Gardner, E.A., 1888. Naukratis. Part II, Sixth Memoir of the Egypt Exploration Fund, London. Ghorbani, M., 2013. The Economic Geology of Iran: Mineral Deposits and Natural Resources, Dordrecht. Giumlia-Mair, A., 2015. Phoenician metalwork: composition and techniques. In: Jiménez Ávila, J. (Ed.), Phoenician bronzes in Mediterranean, Bibliotheca Archaeologica Hisopania 45, Madrid, pp. 481–515. Given, M., Corley, H., Sollars, L., 2007. Joining the dots: continuous survey, routine practice and the interpretation of a Cypriot landscape (with interactive GIS and integrated data archive). Internet Archaeology 20. https://doi.org/10.11141/ia.20.4. Given, M., Knapp, A.B., 2003. The Sydney Cyprus Survey Project, Social Approaches to Social Archaeological Survey, Monumenta Archaeologica 21, Los Angeles. Gjerstad, E., 1948. The Swedish Cyprus Expedition. Finds and Results of the Excavations in Cyprus. Vol. 4 (Stockholm). Gmelins Handbuch der Anorganischen Chemie, 1971. Zinn, Teil A, Verlag Chemie, Weinheim. pp. 315–325. Gorton, A.F., 1996. Egyptian and Egyptianizing Scarabs: A Typology of Steatite, Faience, and Paste Scarabs from Punic and Other Mediterranean Sites, Oxford. Greaves, A.M., 2007. Milesians in the Black Sea: trade, settlement and religion. In: Gabrielson, V., Lund, J. (Eds.), The Black Sea in Antiquity: Regional and Interregional Economic Exchanges, Black Sea Studies 6, Sandbjerg, pp. 9–22. Gubel, E., Cauet, S., 1987. Un nouveau type de Coupe Phénicienne. Syria 64 (3–4), 193–204. Guichard, S., Pierrat-Bonnefois, G., 2005. En Égypte à la Basse Époque. In: Caubet, A., Pierrat-Bonnefois, G. (Eds.), Faïences de l'Antiquité. De l'Égypte à l'Iran, Catalogue de L'exposition du Musée du Louvre 10 Juin–12 Septembre 2005, Paris. Hall, E.T., 1961. Surface-enrichment of buried metals. Archaeometry 4 (1961), 62–66. Hallmann, A., 2009. Was Ancient Egyptian Kohl a Poison? In: Popielska-Grzybowska, J., Białostocka, O., Iwaszczuk, J. (Eds.), Proceeding of the Third Central European Conference of Young Egyptologists. Egypt 2004: Perspectives of Research. Warsaw 12–14 May 2004, Pułtuskpp. 69–72. Hauptmann, A., 2007. The Early Metallurgy of Copper. Evidence from Faynan, Jordan. Natural Science in Archaeology. Heidelberg, Berlin; New-York. Hauptmann, A., Begemann, F., Heitkemper, E., Pernicka, E., Schmitt-Strecker, S., 1992. Early copper produced at Feinan, Wadi Araba, Jordan: the composition of ores and copper. Archeomaterials 6, 1–33. Hauptmann, A., Gambaschidze, I., 2001. Antimon – eine metallurgische Besonderheit aus dem Kaukasus. In: Gambaschidze, I., Hauptmann, A., Slotta, R., Yalçin, Ü. (Eds.), Georgien. Schätze aus dem Land des Goldenen Vlies, Bochum. Hauptmann, A., Maddin, R., Prange, M., 2002. On the structure and composition of copper and tin ingots excavated from the shipwreck of Uluburun. Bull. Am. Sch. Orient. Res. (328), 1–30. Heinz, S., 2015. The production and circulation of metal statuettes and amulets at ThonisHeracleion. In: Robinson, D., Goddio, F. (Eds.), Thonis-Heracleion in Context, Oxford Centre for Maritime Archaeology Monograph 8, Oxford, pp. 55–69. Hill, M., Schorsch, D., 2007. Gifts for the Gods: Images from Egyptian Temples, New York. Höckmann, U., 2007. Zyprisch-griechische Kleinplastik: Kouroi, andere Figuren und plastisch verzierte Gefäße. In: Höckmann, U., Koenigs, W. (Eds.), Archäologische Loeben, Ch.E., Raue, D., Wallenstein, U. (Eds.), Gegossene Götter: Metallhandwerk und Massenproduktion im Alten Ägypten. Marie Leidorf, Rahden, pp. 113–126. Auenmüller, J., Fitzenreiter, M., 2014. Eine Gusswerkstatt auf der Qubbet el-Hawa? In: Fitzenreiter, M., Loeben, Ch.E., Raue, D., Wallenstein, U. (Eds.), Gegossene Götter: Metallhandwerk und Massenproduktion im Alten Ägypten. Marie Leidorf, Rahden, pp. 101–106. Baitinger, H., 2001. Die Angriffswaffen aus Olympia, Berlin and New York. Baralis, A., Panayotova, K., Bogdanova, T., Gyuzelev, M., Nedev, D., Gospodinov, K., 2015. Apollonia Pontica (Sozopol, Bulgaria): the results of the Franco-Bulgarian archaeological mission. In: Manoledakis, M. (Ed.), The Black Sea in the Light of New Archaeological Data and Theoretical Approaches: Proceedings of the 2nd International Workshop on the Black Sea in Antiquity Held in Thessaloniki, 18–20 September 2015, Oxford, pp. 153–179. Baralis, A., Panayotova, K., Bogdanova, T., Gyuzelev, M., Nedev, D., Gospodinov, K., 2016. Apollonia Pontica (Sozopol, Bulgaria): the results of the Franco-Bulgarian archaeological mission. In: Manoledakis, M. (Ed.), The Black Sea in the Light of New Archaeological Data and Theoretical Approaches, Oxford, pp. 153–179. Bartelheim, M., Behrendt, S., Kızılduman, B., Müller, U., Pernicka, E., 2011. Der Schatz auf dem Königshügel, Kaleburnu/Galinoporni, Zypern. Ein Zeugnis internationalen bronzezeitlichen Seehandels im östlichen Mittelmeer. Anatolia 5, 91–110. Bartelheim, M., Kızılduman, B., Müller, U., Pernicka, E., Tekel, H., 2008. The Late Bronze Age Hoard of Kaleburnu/Galinoporni on Cyprus. Památky Archeologické 99, 161–188. Begemann, F., Schmitt-Strecker, S., Pernicka, E., Lo Schiavo, F., 2001. Chemical Composition and Lead Isotopy of Copper and Bronze from Nuragic Sardinia. Eur. J. Archaeol. 4 (1), 43–85. Begemann, F., Hauptmann, A., Schmitt-Strecker, S., Weisgerber, G., 2010. Lead isotope and chemical signature of copper from Oman and its occurrence in Mesopotamia and sites on the Arabian Gulf coast. Arab. Archaeol. Epigr. 21, 135–169. Begemann, F., Schmitt-Strecker, S., Pernicka, E., 2003. On the composition and provenance of metal finds from Beşiktepe (Troia). In: Wagner, G.A., Pernicka, E., Uerpmann, H.-P. (Eds.), Troia and the Troad: Scientific Approaches, Berlin, Heidelberg and New York, pp. 173–201. Bell, L., 2011. A collection of Egyptian bronzes. In: Stager, L.E., Schloen, J.D., Master, D.M. (Eds.), Ashkelon 3: The Seventh Century B.C., Final Reports of The Leon Levy Expedition to Ashkelon. vol. 3. pp. 397–420. Ben-Yosef, E., Shaar, R., Tauxe, L., Levy, T.E., Kassianidou, V., 2011. The Cyprus Archaeomagnetic Project (CAMP): targeting the slag deposits of Cyprus and the Eastern Mediterranean. Antiquity 85 (330). http://antiquity.ac.uk/projgall/benyosef330/. Ben-Yosef, E., Shaar, R., Tauxe, L., Ron, H., 2012. A new chronological framework for lron Age copper production at Timna (Israel). Bulletin of the American School of Oriental Research (367), 31–71. Bernardini, P., Botto, M., 2015. The “Phoenician” bronzes from the Italian Peninsula and Sardinia. In: Jiménez Ávila, J. (Ed.), Phoenician Bronzes in Mediterranean, Bibliotheca Archaeologica Hisopania 45, Madrid, pp. 295–373. Birney, K.J., Levine, E., 2011. Balance weights. In: Stager, L.E., Schloen, J.D., Master, D.M. (Eds.), Ashkelon 3: The Seventh Century B.C., Final Reports of The Leon Levy Expedition to Ashkelon. vol. 3. pp. 473–492. Bokhari, F.Y., Kramers, J.D., 1982. Lead isotope data from massive sulfide deposits in the Saudi Arabian Shield. Econ. Geol. 77, 1766–1769. von Bomhard, A.S. 2015, ‘The stela of Thonis-Heracleion. Economic, topographic and epigraphic aspects’, in D. Robinson and F. Goddio (eds), Thonis-Heracleion in Context, Oxford Centre for Maritime Archaeology Monograph 8, Oxford, 101–120. Carbillet, A., 2011. La Figure Hathorique à Chypre (IIe-Ier Mill. av. J.-C.), AOAT 388, Ugarit-Verlag, Münster. Caubet, A., Pierrat-Bonnefois, G., 2005. Faïences de l'Antiquité. De l'Égypte à l'Iran, Catalogue de l'exposition du Musée du Louvre 10 Juin–12 Septembre 2005, Paris. Charalambous, A., Kassianidou, V., Papasavvas, G., 2014. A compositional study of Cypriot bronzes dating to the Early Iron Age using portable X-ray fluorescence spectrometry (pXRF). J. Archaeol. Sci. 46, 205–216. Chavanne, M.-J., 1990. La Nécropole d'Amathonte. Tombs 110–385. IV. Les Petits Objets, Études Chypriotes XII, Nicosia. Chernykh, E., 1992. Ancient Metallurgy in the USSR, Cambridge. Constantinou, G., 1982. Geological features and ancient exploitation of the cupriferous sulphide orebodies of Cyprus. In: Muhly, J.D., Maddin, R., Karageorghis, V. (Eds.), Early Metallurgy in Cyprus, 4000–500 B.C., Nicosia, pp. 13–23. Courcier, A., 2014. Ancient metallurgy in the Caucasus from the sixth to the third millennium BCE. In: Roberts, B.W., Thornton, C. (Eds.), Archaeometallurgy in Global Perspective: Methods and Syntheses, New York, pp. 579–664. Cour-Marty, M.-A., 1990. Les poids égyptiens, de précieux jalons archéologiques. Cahier de Recherches de l'Institut de Papyrologie et d'Egyptologie de Lille 12, 17–55. Craddock, P.T., Buck, C., 1993. The composition of the mirrors. In: Nicholls, R.V. (Ed.), Corpus Speculorum Etruscorum: Great Britain 2, Cambridge, pp. 53–56. Criss, J.W., LaVerne, S., Birks, L.S., 1968. Calculation methods for fluorescent X-ray spectrometry. Empirical coefficients vs. fundamental parameters. Anal. Chem. 40, 1080–1086. Davies, S., 2007. Bronzes from the Sacred Animal Necropolis at North Saqqara. In: Marsha, H., Schorsch, D. (Eds.), Gifts for the Gods: Images from Egyptian Temples. Metropolitan Museum of Art, New York, pp. 174–187. Davies, W.V., Cowell, R.M., 1987. Catalogue of Egyptian Antiquities in the British Museum. VII, Tools and Weapons. 1, Axes. London. de Callataÿ, F., 2005. The Graeco-Roman economy in the super long-run: lead, copper, and shipwrecks. Journal of Roman Archaeology 18 (1), 261–272. Demetriou, D.A., 2012. Negotiating Identity in the Ancient Mediterranean: The Archaic and Classical Greek Multiethnic Emporia. Cambridge. . 337 Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. Studien zu Naukratis I, Möhnesee. Niederschlag, E., Pernicka, E., Seifert, Th., Bartelheim, M., 2003. Determination of lead isotope ratios by multiple collector ICP-MS: a case study of Early Bronze Age artefacts and their possible relation with ore deposits of the Erzgebirge. Archaeometry 45, 61–100. Nørgaard, H.W., 2017. Portable XRF on prehistoric bronze artefacts: limitations and use for the detection of Bronze Age metal workshops. Open Archaeology 3, 101–122. O'Brien, W., 2015. Prehistoric Copper Mining in Europe: 5500–500 BC, Oxford. Ogden, J., 2000. Metals. In: Shaw, I., Nicholson, P. (Eds.), Ancient Egyptian Materials and Technology, Cambridge, pp. 148–176. Oren, E.D., 1984. Migdol: a new fortress on the edge of the Eastern Nile Delta. Bull. Am. Sch. Orient. Res. (256), 7–44. Pernicka, E., 1987. Erzlagerstätten in der Ägäis und ihre Ausbeutung im Altertum: Geochemische Untersuchungen zur Herkunftsbestimmung archäologischer Metallobjekte. Jahrb. Röm.-Germ. Zentralmus 34, 607–714. Pernicka, E., 2014. Provenance determination of archaeological metal objects. In: Roberts, B.W., Thornton, C. (Eds.), Archaeometallurgy in Global Perspective. Methods and Syntheses. Springer, New York, pp. 239–268. Pernicka, E., Adam, K., Böhme, M., Hezarkhani, Z., Nezafati, N., Schreiner, M., Winterholler, B., Momenzadeh, M., Vatandoust, A., 2011. Archaeometallurgical researches at Arisman in central Iran. In: Vatandoust, A., Parzinger, H., Helwing, B. (Eds.), Early Mining and Metallurgy on the Central Iranian Plateau. Report on the First Five Years of Research of the Joint Iranian-German Research Project. Archäologie in Iran und Turan 9, Mainz am Rhein, pp. 633–705. Pernicka, E., Begemann, F., Schmitt-Strecker, S., Grimanis, A.P., 1990. On the composition and provenance of metal artefacts from Poliochni on Lemnos. Oxf. J. Archaeol. 9 (3), 263–297. Pernicka, E., Seeliger, T.C., Wagner, G.A., Begemann, F., Schmitt-Strecker, S., Eibner, C., Öztunalı, Ö., Baranyi, I., 1984. Archäometallurgische Untersuchungen in Nordwestanatolien. Jahrb. Röm.-Germ. Zentralmus 31, 533–599. Petrie, W.M.F., 1886. Naukratis. Part I., 1884-5, Third Memoir of the Egypt Exploration Fund, London. Petrie, W.M.F., 1888. Tanis 2: Nebesheh (Am) and Defenneh (Tahpanhes), London. Pichot, V., Fluzin, P., Valloggia, M., Wuttmann, M., 2006. Les chaînes opératoires métallurgiques en Égypte à l'époque gréco-romaine: premiers résultats archéométriques et archéologiques. In: Mathieu, B., Meeks, D., Wissa, M. (Eds.), L'apport de l'Égypte à L'histoire des Techniques, Bibliothèque d'Étude 142, Cairo, pp. 217–238. Pike, A., 2002. Analysis of Caucasian metalwork – the use of antimonal, arsenical and tin bronze in the Late Bronze Age. In: Curtis, J., Kruszyński, M. (Eds.), Ancient Caucasian and Related Material in The British Museum, British Museum Occasional Paper 121, London, pp. 87–92. Prange, M.K., 2001. 5000 Jahre Kupfer in Oman. Band II. Vergleichende Untersuchungen zur Charakterisierung des omanischen Kupfers mittels chemischer und isotopischer Analysenmethoden. Metalla (Bochum) 8, 1–126. Rademakers, Fr.W., Rehren, T.H., Pernicka, E., 2017. Copper for the Pharaoh: identifying multiple metal sources for Ramesses' workshops from bronze and crucible remains. J. Archaeol. Sci. 80, 50–73. Razmjou, S., Tite, M.S., Shortland, A.J., Jung, M., Hauptman, A., 2004. Glazed bricks in the Achaemenid Period. In: Stöllner, T., Slotta, R., Vatandoust, R. (Eds.), Persiens Antike Pracht. Bergbau-Handwerk-Archäologie. Katalog der Ausstellung des Deutschen-Bergbau-Museums Bochum 2004–2005, Bochum, pp. 382–393. Rehren, Th., Boscher, L., Pernicka, E., 2012. Large scale smelting of speiss and arsenical copper at Early Bronze Age Arisman, Iran. J. Archaeol. Sci. 39 (6), 1717–1727. Reyes, A.T., 1992. The anthropomorphic statuettes of archaic Idalion. In: Annual of the British School at Athens 87, London and Athenspp. 243–257. Riederer, J., 1981. Analysis of Egyptian bronzes. Revue d'Archéométrie 1, 239–243. Riederer, J., 1988. Metallanalysen ägyptischer Bronzestatuetten aus deutschen Museen. Berliner Beiträge zur Archäometrie 10, 5–20. Robinson, D., 2016. In Goddio and Masson-Berghoff 2016, London. pp. 112. Robinson, Z., 2010. Living with metals in Hellenistic Egypt – new finds from HeracleionThonis. In: Robinson, D., Wilson, A. (Eds.), Alexandria and the North-West Delta, Oxford, pp. 219–231. Roebuck, C., 1950. The Grain Trade between Greece and Egypt. Class. Philol. 45 (4), 236–247. Roeder, G., 1937. Ägyptische Bronzewerke. Pelizaeus Museum zu Hildesheim, Wissenschaftliche Veröffentlichung 3, Glückstadt. Roeder, G., 1956. Ägyptische Bronzefiguren, Mitteilungen aus der Ägyptischen Sammlung, Staatliche Museen zu Berlin 6, Berlin. Sayre, E.V., Yener, K.A., Joel, E.C., Barnes, I.L., 1992. Statistical evaluation of the presently accumulated lead isotope data from Anatolia and surrounding regions. Archaeometry 34, 73–105. Schorsch, D., 1988. In: Watkins, S.C., Brown, C.E. (Eds.), Technical Examinations of Ancient Egyptian Theriomorphic Hollow Cast Bronzes—Some Case Studies. Conservation of Ancient Egyptian Materials, Londonpp. 41–50. Schorsch, D. forthcoming, ‘Ritual metal statuary in ancient Egypt: “A long life and a great and good old age”’, in A. Masson-Berghoff (eds), Statues in context: production, meaning and (re)uses. Proceedings of the Sackler colloquium held in London, 13-14 July 2016. Schulze, M., Lehmann, R., 2014. Alte und neue Analyseverfahren an den Bronzen des Museums August Kestner und was sie uns verraten. In: Fitzenreiter, M., Loeben, C.E., Raue, D., Wallenstein, U. (Eds.), Gegossene Götter. Metallhandwerk und Massenproduktion im Alten Ägypten, Rahden/Westf. pp. 133–154. Schwab, R., Willer, F., 2016. Die Zusammensetzung und mögliche Herkunft der Gusslegierungen. In: Fitzenreiter, M., Willer, F., Auenmüller, J. (Eds.), Materialien einer Gusswerkstatt von der Qubbet el-Hawa, Bonner Sammlung von Aegyptiaca 7, Studien zu Naukratis II, Worms, pp. 13–307. Jiménez Ávila, J., 2015. ‘Figuras Fenicias del Mediterráneo. Caracterización y novedades’. In: Jiménez Ávila, J. (Ed.), Phoenician bronzes in Mediterranean, Bibliotheca Archaeologica Hispana. 45. pp. 197–230 Madrid. Johnston, A.W., 1982. Fragmenta Britannica II: sherds from Naukratis. 3. Cypriot amphorae and 4. Etruscan Bucchero. Bulletin of the Institute of Classical Studies 29, 35–42. Karageorghis, V., Raptou, E., 2014. Necropoleis at Palaepaphos From the End of the Late Bronze Age to the Cypro-Archaic Period, Nicosia. Kassianidou, V., 2012. The origin and use of metals in Iron Age Cyprus. In: Iacovou, M. (Ed.), Cyprus and the Aegean in the Early Iron Age. The Legacy of Nicolas Coldstream, pp. 229–259. Kassianidou, V., 2013. The exploitation of the landscape: metal resources and the copper trade during the age of the Cypriot city-kingdoms. Bulletin of the American School of Oriental Research (370), 49–82. Kiderlen, M. and Bode, M. forthcoming, ‘Greek copper markets: Modelling the imports to Greek central places ca. 1100 – 700 BC’, in Proceedings of the 19th International Congress of Classical Archaeology in Cologne and Bonn, 22 – 26 May 2018. Kiderlen, M., Bode, M., Hauptmann, A., Bassiakos, Y., 2016. Tripod cauldrons produced at Olympia give evidence for trade with copper from Faynan (Jordan) to South West Greece, c. 950–750 BCE. J. Archaeol. Sci. Rep. 8, 303–313. Kiely, T., 2014. Statuette of worshipper in Egyptian-style dress. In: Aruz, J., Graff, S., Rakic, Y. (Eds.), From Assyria to Iberia: At the Daw of Classical Age, New Haven, pp. 196–197. Koenigs, W., 2007. Archaische griechische Bauteile. In: Höckmann, U., Koenigs, W. (Eds.), Archäologische Studien zu Naukratis II, Worms, pp. 311–384. Kroll, J.H., 2003. Weights, bullion, currency, coinage. In: Stampolidis, N.Ch., Karageorghis, V. (Eds.), Ploes. Sea Routes… Interconnections in the Mediterranean, c. 1600–600 BC. Procedings of the International Symposium Held in Rethymnon, Crete September 29th-October 2, 2002, Athenspp. 313–323. Leclère, F., Spencer, J., 2014. Tell Dafana reconsidered. The archaeology of an Egyptian frontier town. British Museum Research Publication 199. Levy, T.E., Ben-Yosef, E., Najjar, M., 2012. New perspectives on Iron Age copper production and society in the Faynan region, Jordan. In: Kassianidou, V., Papasavvas, G. (Eds.), Eastern Mediterranean Metallurgy and Metalwork in the Second Millennium BC - A Conference in Honour of James D. Muhly, Nicosia, 10th–11th October 2009, Oxford, pp. 197–214. Liu, S., Rehren, Th., Pernicka, E., Hausleiter, A., 2015. Copper processing in the oases of northwest Arabia: technology, alloys and provenance. J. Archaeol. Sci. 53, 492–503. Lutz, J., Pernicka, E., 1996. Energy dispersive X-ray fluorescence analysis of ancient copper alloys: empirical values for precision and accuracy. Archaeometry 38, 313–323. Masson, A., 2014. New Year's flask. In: Villing et al. 2013–19, [last accessed: 24/07/ 2018]. http://www.britishmuseum.org/research/online_research_catalogues/ng/ naukratis_greeks_in_egypt/material_culture_of_naukratis/new_years_flasks.aspx. Masson, A., 2015a. Cult and trade. A reflexion on Egyptian metal offerings from Naukratis. In: Robinson, D., Goddio, F. (Eds.), Thonis-Heracleion in context, Oxford Centre for Maritime Archaeology Monograph 8, Oxford, pp. 71–88. Masson, A., 2015b. Bronze votive offerings. In: Villing et al. 2013–19, [last accessed: 24/ 07/2018]. http://www.britishmuseum.org/research/online_research_catalogues/ ng/naukratis_greeks_in_egypt/material_culture_of_naukratis/bronze_offerings.aspx. Masson, A., 2015c. Foundation deposits. In: Villing et al. 2013–19, [last accessed: 24/07/ 2018]. http://www.britishmuseum.org/research/online_research_catalogues/ng/ naukratis_greeks_in_egypt/material_culture_of_naukratis/foundation_deposits.aspx. Masson, A., 2016. Interpréter le matériel grec et chypriote dans un contexte religieux et thébain: l'exemple du quartier des prêtres de Karnak – des consommateurs égyptiens de produits grecs et chypriotes. In: Gorre, G., Marangou, A. (Eds.), La Présence Grecque Dans la Vallée de Thèbes, Rennes, pp. 25–43. Masson, A., 2018a. Scarabs, scaraboids and amulets. In: Villing et al. 2013–19, [last accessed: 24/07/2018]. http://www.britishmuseum.org/research/online_research_ catalogues/ng/naukratis_greeks_in_egypt/material_culture_of_naukratis/scarabs_ scaraboids_and_amulets.aspx. Masson, A., 2018b. Seals and seal impressions. In: Villing et al. 2013–19, . http://www. britishmuseum.org/research/online_research_catalogues/ng/naukratis_greeks_in_ egypt/material_culture_of_naukratis/seals_and_seal_impressions.aspx. Masson, A. in preparation, ‘Weights’, In: Villing et al. 2013–19. Masson, O., 1968. Kypriaka VII: Le sanctuaire d'Apollon à Idalion (Fouilles 1868–1869). BCH 92, 386–402. Matthäus, H., 1985. Metallgefäße und Gefäßuntersätze der Bronzezeit, der Geometrischen und Archaischen Periode auf Zypern. Prähistorische Bronzefunde Abteilung II 8, Munich. McConnell, J.R., Wilson, A.I., Stohl, A., Arienzo, M.M., Chellman, N.J., Eckhardt, S., Thompson, E.M., Pollard, A.M., Peder Steffensen, J., 2018. Lead pollution recorded in Greenland ice indicates European emissions tracked plagues, wars, and imperial expansion during antiquity. In: Proceedings of the National Academy of Sciences May 2018. Meek, A., Bouquillon, A., Lehuédé, P., Masson, A., Villing, A., Pierrat-Bonnefois, G., Webb, V., 2016. Discerning differences: ion beam analysis of ancient faience from Naukratis and Rhodes. Une Europe de la Recherche en Sciences du Patrimoine, Technè 43 (2016), 95–101. Möller, A., 2000. Naukratis: Trade in Archaic Greece, Oxford. Moreno, A., 2007. Feeding the Democracy: The Athenian Grain Supply in the Fifth and Fourth Centuries BC, Oxford. Murray, A.S., Smith, A.H., Walters, H.B., 1900. Excavations in Cyprus: bequest of Miss E.T. Turner to the British Museum, London. Nick, G., 2006. Zypro-Ionische Kleinplastik aus Kalkstein und Alabaster, Archäologische 338 Journal of Archaeological Science: Reports 21 (2018) 318–339 A. Masson-Berghoff et al. Vacek, A., 2017. Al Mina and changing patterns of trade: the evidence from the eastern Mediterranean. In: Charalambidou, X., Morgan, C. (Eds.), Interpreting the Seventh Century BC. Tradition, Innovation and Meaning. Oxford, pp. 47–59. Vaelske, V., Bode, M. and el Morr, Z. forthcoming, ‘Early Iron Age Copper Trails: The Case of Phoenicia. First Results of a Pilot-study’, submitted to: Bulletin d’archéologie et d’architecture libanaises. van Alfen, P.G., 2002. Pant'agatha: Commodities in Levantine-Aegean Trade During the Persian Period, 6–4th c. BC. PhD thesis. The University of Texas at Austin. van der Wilt, E., 2010. In: Hudecz, A., Petrik, M. (Eds.), Lead Weights and Ingots from Heracleion-Thonis: an Illustration of Egyptian Trade Relations with the Aegean. Commerce and Economy in Ancient Egypt, BAR International Series 2131, Oxfordpp. 157–164. van der Wilt, E., 2014. Large lead containers from Thonis-Heracleion, Egypt: metal storage vessels? Anatolica 60, 219–231. Vavelidis, M., Gialoglou, G., Pernicka, E., Wagner, G.A., 1988. Die Buntmetall- und EisenMangan-Lagerstätten von Thasos. In: Wagner, G.A., Weisgerber, G. (Eds.), Antike Edel- und Buntmetallgewinnung auf Thasos. Der Anschnitt, Beiheft_6, Deutsches Bergbau-Museum, Bochum, pp. 40–58. Vikentiev, V., 1957. Les trois inscriptions concernant la mine de plomb d'Oum Huetat. Annales du Service des Antiquités de l'Égypte 54, 179–189. Villing, A., 2006. ‘Drab Bowls’ for Apollo: the Mortaria of Naukratis and Exchange in the Archaic Eastern Mediterranean. In: Villing, A., Schlotzhauer, U. (Eds.), Naukratis: Greek Diversity in Egypt, Studies on East Greek Pottery and Exchange in the Eastern Mediterranean, London, pp. 31–46. Villing, A., 2013a. Egypt as a “Market” for Greek Pottery: some thoughts on production, consumption and distribution in an intercultural environment. In: Tsingarida, A., Viviers, D. (Eds.), Pottery Markets in the Ancient Greek World (8th–1st Centuries B.C.), Études d'Archéologie 5, Brussels, pp. 73–102. Villing, A., 2013b. Naukratis: a city and trading port in Egypt. In: Villing et al. 2013–19, [last accessed: 24/07/2018]. http://www.britishmuseum.org/research/online_ research_catalogues/ng/naukratis_greeks_in_egypt/introduction/naukratis_a_city_ and_port.aspx. Villing, A., 2015. Egyptian-Greek exchange in the Late Period: the view from NokradjNaukratis. In: Robinson, D., Goddio, F. (Eds.), Thonis-Heracleion in Context, Oxford Centre for Maritime Archaeology Monograph 8, Oxford, pp. 229–246. Villing, A., Schlotzhauer, U. (Eds.), 2006. Naukratis: Greek Diversity in Egypt. Studies on Greek Pottery and Exchange in the Eastern Mediterranean. British Museum Research Publication 162, London. Villing, A., Bergeron, M., Bourogiannis, G., Johnston, A., Leclère, F., Masson, A., Thomas, R., 2013–19. Naukratis: Greeks in Egypt, British Museum Online Research Catalogue. [last accessed: 24/07/2018]. http://www.britishmuseum.org/research/online_ research_catalogues/ng/naukratis_greeks_in_egypt.aspx. Vogl, J., Rosner, M., Curbera, J., Peltz, U., Peplinski, B., 2016. Lead isotope analysis in magic artefacts from the Berlin museums. Archaeol. Anthropol. Sci. 1–17. Vonhoff, C., 2015. Phoenician bronzes in Cyprus. In: Jiménez Ávila, J. (Ed.), Phoenician Bronzes in Mediterranean, Bibliotheca Archaeologica Hisopania 45, Madrid, pp. 269–296. Wagner, G.A., Begemann, F., Eibner, C., Lutz, J., Öztunalı, Ö., Pernicka, E., SchmittStrecker, S., 1989. Archäometallurgische Untersuchungen an Rohstoffquellen des frühen Kupfers in Ostanatoliens. Jahrb. Röm.-Germ. Zentralmus 36, 637–686. Wagner, G.A., Öztunalı, Ö., 2000. Prehistoric copper sources in Turkey. In: Yalçın, Ü. (Ed.), Anatolian Metal I. Der Anschnitt, Beiheft 13, Selbstverlag des Deutschen Bergbau-Museums, Bochum, pp. 31–67. Wagner, G.A., Pernicka, E., Seeliger, T.C., Lorenz, I.B., Begemann, F., Schmitt-Strecker, S., Eibner, C., Oztunalı, Ö., 1986. Geochemische und isotopische Charakteristika früher Rohstoffquellen für Kupfer, Blei, Silber und Gold in der Türkei, Jahrb. Röm.-Germ. Zentralmus 33, 723–730. Wagner, G.A., Pernicka, E., Seeliger, T.C., Oztunalı, Ö, Baranyi, I., Begemann, F., SchmittStrecker, S., 1985. Geologische Untersuchungen zur frühen Metallurgie in NWAnatolien. Bull. Min. Res. Expl. Inst. Turkey 101 (/102), 45–81. Weeks, L., Keall, E., Pashley, V., Evans, J., Stock, S., 2009. Lead isotope analyses of Bronze Age copper-base artefacts from al-Midamman, Yemen: towards the identification of an indigenous metal production and exchange system in the southern Red Sea region. Archaeometry 51, 576–597. Weinstein, J.M., 1973. Foundation Deposits in Ancient Egypt, Ann Arbor. Weisgerber, G., 2006. The mineral wealth of ancient Arabia and its use I: copper mining and smelting at Feinan and Timna - comparison and evaluation of techniques, production, and strategies. Arab. Archaeol. Epigr. 17 (1), 1–30. Weiss, K., 2012. Ägyptische Tier- und Götterbronzen aus Unterägypten: Untersuchungen zu Typus, Ikonographie und Funktion sowie der Bedeutung innerhalb der Kulturkontakte zu Griechenland, Ägypten und Altes Testament 81, Wiesbaden. Wuttmann, M., Coulon, L., Gombert, F., 2008. An assemblage of bronze statuettes in a cult context: the temple of ‘Ayn Manawîr’. In: Hill, M. (Ed.), Gifts of the Gods, Images from Egyptian Temples, New York, pp. 167–173. Yahalom-Mack, N., Galili, E., Segal, I., Eliyahu-Behar, A., Boaretto, E., Shilstein, S., Finkelstein, I., 2014. New insights into Levantine copper trade: analysis of ingots from the Bronze and Iron Ages in Israel. J. Archaeol. Sci. 45, 159–177. Yener, K.A., Sayre, E.V., Joel, E.C., Özbal, H., Barnes, I.L., Brill, R.H., 1991. Stable lead isotope studies of central Taurus ore sources and related artifacts from eastern Mediterranean Chalcolithic and Bronze Age sites. J. Archaeol. Sci. 18, 1–37. Berlin, pp. 71–81. Seeliger, T.C., Pernicka, E., Wagner, G.A., Begemann, F., Schmitt-Strecker, S., Eibner, C., Öztunalı, Ö., Baranyi, I., 1985. Archäometallurgische Untersuchungen in Nord- und Ostanatolien. Jahrb. Röm.-Germ. Zentralmus. 32, 597–659. Senff, R., 1993. Das Apollonheiligtum von Idalion: Architektur und Statuenausstattung eines zyprischen Heiligtums, Studies in Mediterranean archaeology XCIV, Jonsered. Shortland, A.J., 2002. The use and origin of antimonate colorants in early Egyptian glass. Archaeometry 44, 517–530. Shortland, A.J., 2006. Application of lead isotope analysis to a wide range of Late Bronze Age Egyptian materials. Archaeometry 48 (4), 657–669. Shortland, A.J., 2009. The implications of lead isotope analysis for the source of pigments in Late Bronze Age Egyptian vitreous materials. In: Degryse, P., Henderson, J., Hodgins, G. (Eds.), Isotopes in Vitreous Materials, Studies in Archaeological Sciences 1, Leuven, pp. 99–112. Shortland, A.J., 2012. Lapis Lazuli from the Kiln: Glass and Glassmaking in the Late Bronze Age, Leuven. Stacey, J.S., Doe, B.R., Roberts, R.J., Delevaux, M.H., Gramlich, J.W., 1980. A lead isotope study of mineralization in the Saudi Arabian Shield. Contrib. Mineral. Petrol. 74, 175–188. Stos, Z.A., 2009. Across the wine dark seas… sailor, tinkers and royal cargoes in the Late Bronze Age eastern Mediterranean. In: Shortland, A.J., Freestone, I., Rehren, Th. (Eds.), From Mine to Microscope. Advances in the Study of Ancient Technology, Oxford, pp. 163–180. Stos-Gale, Z.A., 2000. Trade in metals in the Bronze Age Mediterranean: an overview of lead isotope data for provenance studies. In: Pare, C. (Ed.), Metals Make the World Go Round: Supply and Circulation of Metals in Bronze Age Europe, Oxford, pp. 56–69. Stos-Gale, Z.A., 2001. The impact of the natural sciences on studies of Hacksilber and early silver coinage. In: Balmuth, M.S. (Ed.), Hacksilber to Coinage: New Insights into the Monetary History of the Near East and Greece. Collection of Eight Papers Presented at the 99th Annual Meeting of the Archaeological Institute of America, Numismatic Studies 24, New York, pp. 53–76. Stos-Gale, Z.A., Gale, N.H., 1981. Sources of galena, lead and silver in Predynastic Egypt, Actes du XXème Symposium International d'Archéometrie. Revue d'Archéometrie 5, 285–295. Stos-Gale, Z.A., Gale, N.H., Annetts, N., 1996. Lead isotope analyses of ores from the Aegean. Archaeometry 38 (2), 381–390. Stos-Gale, Z.A., Gale, N.H., Houghton, J., Speakman, R., 1995. Lead isotope data from the Isotrace laboratory, Oxford; Archaeometry data base 1, ores from the western Mediterranean. Archaeometry 37, 407–415. Stos-Gale, Z.A., Maliotis, G., Gale, N.H., Annetts, N., 1997. Lead isotope characteristics of the Cyprus copper ore deposits applied to provenance studies of copper oxhide ingots. Archaeometry 39 (1), 83–123. Swaddling, J., 2001. Corpus speculorum Etruscorum. Great Britain. 1, British Museum. Fasc. 1, Archaic and Classical Tanged Mirrors (Early Tanged & Related Mirrors), London. Tadmor, M., Kedem, D., Begemann, F., Hauptmann, A., Pernicka, E., Schmitt-Strecker, S., 1995. The Nahal Mishmar hoard from the Judean Desert: technology, composition, provenance. Atiqot 27, 93–146. Tafel, V., Wagenmann, K., 1951. Lehrbuch der Metallhüttenkunde I, Leipzig. Thomas, R.I., 2015a. Naukratis, “Mistress of ships” in context. In: Robinson, D., Goddio, F. (Eds.), Thonis-Heracleion in context, Oxford Centre for Maritime Archaeology Monograph 8, Oxford, pp. 247–265. Thomas, R.I., 2015b. Cypriot figures in terracotta and limestone. In: Villing et al. 2013–19, [last accessed: 24/07/2018]. http://www.britishmuseum.org/research/ online_research_catalogues/ng/naukratis_greeks_in_egypt/material_culture_of_ naukratis/cypriot_figures.aspx. Thomas, R.I., 2017. Tools and weapons. In: Villing et al. 2013–19, [last accessed: 24/07/ 2018]. http://www.britishmuseum.org/research/online_research_catalogues/ng/ naukratis_greeks_in_egypt/material_culture_of_naukratis/tools_and_weapons.aspx. Thomas, R.I., Acosta, C., 2018. Jewellery and mirrors. In: Villing et al. 2013–19, [last accessed: 24/07/2018]. http://www.britishmuseum.org/research/online_research_ catalogues/ng/naukratis_greeks_in_egypt/material_culture_of_naukratis/jewellery_ and_mirrors.aspx. Thornton, C.P., Rehren, Th., Pigott, V.C., 2009. The production of speiss (iron arsenide) during the Early Bronze Age in Iran. J. Archaeol. Sci. 36, 308–316. Tiverios, M., 2008. Greek Colonisation of the Northern Aegean. In: Tsetskhladze, G.R. (Ed.), Greek Colonisation. An Account of Greek Colonies and Other Settlements Overseas II, Leiden, pp. 1–154. Treister, M.Y., 1998. Ionia and the North Pontic Area: Archaic metalworking. In: Tsetskhladze, G. (Ed.), The Greek Colonisation of the Black Sea Area: Historical Interpretation of Archaeology, Wiesbaden. Tsetskhladze, G.R., 1995. Did the Greeks go to Colchis for metals? Oxf. J. Archaeol. 14 (3), 307–331. Tsetskhladze, G.R., 1998. Trade on the Black Sea in the archaic and classical periods: some observations. In: Parkins, H., Smith, C.H. (Eds.), Trade, Traders, and The Ancient City, London and New York, pp. 52–74. Tsetskhladze, G.R., 2016. Greeks in the Asiatic Bosporus: new evidence and some thoughts. In: Manoledakis, M. (Ed.), The Black Sea in the Light of New Archaeological Data and Theoretical Approaches. Proceedings of the 2nd International Workshop on the Black Sea in antiquity held in Thessaloniki, 18–20 September 2015, Oxfordpp. 45–59. 339