Eastern Mediterranean Metallurgy in the Second Millennium BC

By Vasiliki Kassianidou

James D. Muhly is a distinguished scholar with a special interest in ancient metallurgy who has dedicated much of his research to Cypriot archaeology. His work on the metallurgy of ancient Cyprus endorses the true importance of the island as a copper producing region, as well as a pioneer in the development and spread of metallurgy and metalwork in the wider eastern and central Mediterranean region. This volume contains papers from "Eastern Mediterranean Metallurgy and Metalwork in the Second Millennium BC", an international conference organised in Muhly's honour by the University of Cyprus. Several archaeologists and archaeometallurgists from around the world whose research focuses on the metallurgy of this period in Cyprus and surrounding regions were invited to participate in the conference to compare and contrast the material culture associated with metallurgical workshops and to discuss technological issues and their cultural and archaeological contexts. Some papers are devoted to the metallurgy and metalwork of Cyprus, presenting material from various sites and discussing the production and use of copper in the eastern Mediterranean. Others are dedicated to the Minoan and Aegean metal industry and the connections between Sardinia and Cyprus. Moving eastwards, from Anatolia through the Syro-palestinian coast and Jordan and south to Egypt, papers are presented that discuss Late Bronze Age metallurgy in Alalakh, Ugarit, Faynan, Timna and Qantir. The volume also includes papers on tin and iron.

• Language: English
• Publisher: Oxbow Books
• Release date: May 31, 2012
• ISBN: 9781842179574

Book preview

1. Reminiscences: working with Jim Muhly

Robert Maddin

---

The letter I received from Professors Kassianidou and Papasavvas requested that all invited present their most recent (unpublished) studies at a conference following the ceremonies honouring Jim Muhly. While eager to be present at the occasion for one I had worked with for so many years, but well aware of not having done any original scholarship in the field since the studies on the metals from the Uluburun shipwreck (Maddin 1989; Hauptmann et al. 2002), I realized that rehashed studies were not what was wanted. In a telephone conversation with Lina we reached agreement that I could spend a few minutes opening the conference reminiscing about some adventures from those years. This I could easily agree to since there are so many to choose from after 36 years – and they still continue today.

It was in September of 1973 that I first met Jim at the faculty club of the University of Pennsylvania in Philadelphia. I had read an article he had written on trade in metals in the ancient Near East which, as a professor of materials science, had drawn my interest. That interest prompted me to seek out Jim Muhly who, I was pleasantly surprised to learn, was a professor (in Near Eastern Studies) at the same university. Our first lunch and the many that followed convinced me of how my interests and his converged. Those times were spent planning goals for the summer of 1974 and possibly later years. Muhly’s doctoral thesis contained among many other subjects a compilation of the details of published ‘oxhide-shaped’ copper ingots (Muhly 1973). Many intriguing questions were raised; there appeared to be no conventional wisdom concerning the reasons for their shape or their existence. I recall Jim saying that because of a sort of constancy of weight many considered them a form of currency. We decided to sample all known ingots and examine them both for chemical elements and for their metallurgical structure. In preparation I designed a coring tool naively believing that a curator would permit me to desecrate the items for which they were held responsible. That idea was nixed by the Greek customs agents who confiscated the copper plugs I had brought to fill the holes made in coring the ingots; neither Jim nor I could convince them why they were needed. Ironically, a few years ago Cemal Pulak, the director of the Uluburun shipwreck excavation, permitted us to take cores from a number of the copper ingots from that wreck; the metallurgical and mineralogical examination (published a few years ago in BASOR: Hauptman et al. 2002) of those cores yielded valuable information, information that would have obviated many future studies and clarified some confusing character istics of excavated oxhide ingots and fragments. The director of the Greek Numismatic Museum where the ingots were exhibited, perhaps reflecting the view of their being a form of currency, permitted us to take drillings with only the smallest drills found in Athens. Although knowing the quantities of the trace elements is most important, it would have even more interesting to look at the metal-lographic structure.

The next phase of that summer’s plan was to sample the copper ingots from the Cape Gelidonya shipwreck stored in the museum in Bodrum, Turkey. This was in July of 1974; because of the war we had to give up that phase. Since Jim was due in Crete only a few days later, I spent some time in Crete. This was fortunate for there we met an archaeologist who later arranged a meeting with Tamara Stech; she joined our group in late 1974. In the summer of the next year the three of us made it to Bodrum where after a frustrating three days waiting for the director of the museum to return we were able to obtain samples from the ingots, of suitable size for both elemental analysis and metallography. Bodrum in 1975 was a small village, not the trendy exceedingly crowded popular resort with large luxurious hotel complexes it is today. In fact, the motel where we were able to find accommodations, sort of a 1920-type motel typical of road traveling America, appropriately called the Halicarnassus, consisted of a rectangular array of spartan rooms with running water irregularly scheduled about two hours each day.

Over the next few years we continued sampling oxhide ingots from the various museums wherever they could be located. One summer with both Jim and Tammy otherwise busy, I traveled to Sardinia to sample the Serra Illixi ingot in Cagliari, and then to Sicily to sample the Canatello ingot; both attempts were unsuccessful, in Sardinia because of a turf battle and in Sicily because no one could locate the ingot or even say that it existed. Only just a few years ago, Fulvia Lo Schiavo reported its existence; it is now correctly catalogued (see the most recent and comprehensive publication on all the Central Mediterranean ingots in Lo Schiavo et al. (eds) 2009). The results of these studies were published in various journals (see list of references below).

Although Jim had previously worked in Cyprus, my introduction to Cyprus was in 1978 when called upon to consult with Trude Dothan and Amnon Ben-Tor, directors of the excavation at Athienou (Maddin et al. 1983). We returned many times to work on material from Karageorghis’ excavations at Kition and Palaepaphos-Skales, as well as others. (Actually, Andreas Hauptmann and I recently returned to Nicosia to re-sample the curious conglomerate from the metallurgical debris from Kition and some encrustations from the inside of the Enkomi crucible – Hauptmann is publishing these studies in a volume soon to be released). We also worked on material from Ian Todd and Alison South’s excavation at Kalavasos, and materials from many other excavations.

Having often discussed smelting furnaces in Philadelphia, we asked ourselves the question, how high a temperature could be achieved with and without forced air? In the summer of 1978, we set out for Cyprus with thermocouples and temperature measuring equipment in order to answer this question. By arrangement we met George Constantinou and George Malliotis at the Kambia mines in a maintenance shed adjacent to a hill containing a small hollow suitable as a furnace. Gathering copper ores from the nearby open pit mine and using available charcoal, inserting the thermocouple and the charge, we fired up the furnace. The maximum temperature we could obtain was 800°C. After the furnace cooled, we again charged up the furnace and inserted an iron pipe about 2.5cm in diameter into the expected reaction zone with the other the end attached to George Malliotis’ wife’s hair dryer electrically connected to the mains. Firing up the furnace and blowing in the air raised our temperature to 1200°C within minutes. After about ten minutes and cooling the furnace the slag lumps were broken and showed with naked eye large abundant copper prills. These results were never published since there were so many questions we left unanswered. George Constantinou, however, had brought along lamb which we roasted and along with the wine we celebrated well into the afternoon.

Without doubt, prominent among remembrances of Cyprus is the time spent arranging the 1981 Larnaca conference (Early Metallurgy in Cyprus, 4000–500 BC, Muhly et al. 1982). Definite plans were made in 1980 in Nicosia. Quite fortunately while Jim and I were staying at the Cyprus American Archaeological Research Institute (CAARI) and discussing these plans, Fulvia Lo Schiavo was also staying there and introduced herself. She was carrying a black loose-leaf notebook with photographs of some complete oxhide ingots and hundreds of oxhide and plano-convex ingot pieces which she and others had excavated in Sardinia. At the time, as far as I can remember, Sardinia was hardly on the early metallurgy radar screen. This meeting began a relationship that led to our many years of studies of copper ingots in Sardinia, resulting in two books on the early metallurgy of Sardinia (Lo Schiavo et al. (eds) 2005) and the central Mediterranean (Lo Schiavo et al. (eds) 2009), the latter just released.

Arrangements for the Larnaca conference continued with the help and guidance of Vassos Karageorghis; he led us to the Pierides Foundation who graciously funded the conference. The news of the upcoming conference became known almost six months before the scheduled time of June 1981. Many interested scholars and scientists wrote to request permission to attend but, because we were limited to the accommodations available at the hotel and by the funding to fifty invitees, we had to refuse many of those who wanted to come; one of these was Andreas Hauptmann just beginning his career.

In 1971, a Swedish metallurgist, Eric Tholander, published the presence of martensite in a knife blade from Idalion (Idalion 106) on Cyprus with the reported date of c. 1000 BC (Tholander 1971). Briefly, martensite is a unique metallurgical phase that shows up only in carburized iron heated and quenched into a cold medium, e.g. water. Its presence is prima facie evidence that the knife had been carburized and heat treated. We felt that this report had to be verified. So, in 1981, Jim and I set out for Stockholm with permission to sample that knife and other iron objects from Idalion, Amathus and Lapithos (from tombs also excavated by the Swedish Cyprus Expedition). At that time I was Editor-in-Chief of an international journal (Materials Science and Engineering) published by Elsevier in Amsterdam but with its editorial office in Oxford; hence, I was a frequent visitor to Oxford. The editorial office was just up the Banbury Road less than a kilometre from ‘The Old Parsonage’ and within a block of a restaurant ‘Quat’ Saisons’. I considered it the best in Oxford or, for that matter, in England. On returning to Philadelphia I would boast to Jim about that restaurant. On our way to Stockholm, since we were changing planes in Heathrow, we arranged to spend part of the afternoon in Oxford and have lunch at ‘Quat’ Saisons’. On the trip to Oxford Jim announced that he had made a date for us to have lunch with Vassos, then in residence at Merton College of which he is a Fellow. Assured by Jim (as I had also heard) that Merton is considered to have one of the best kitchens of the Oxford colleges, I agreed with some reluctance. We had a wonderful talk with Vassos after which we were served – cold pasta. What a shock! Jim didn’t have a chance to lunch that day at ‘Quat’ Saisons’. A few years later, however, he did have dinner at ‘Le Manoir aux Quat’ Saisons’, the same restaurant having moved to nearby Great Milton. In Stockholm we were able to obtain 15 samples including Idalion 106 and confirm the presence of martensite.

These are just a few reminiscences from 36 years of working with Jim Muhly; what is uppermost in my memories about them all is that he is a great travelling companion both interesting and a joy. It has been a great ride.

Bibliography

Åström, P., Maddin, R., Muhly, J. D. and Stech, T. (1986) Iron artifacts from Swedish excavations in Cyprus. Opuscula Atheniensia 16, 27–41.

Hauptmann, A. (2011) Slags from the Late Bronze Age workshops at Kition and Enkomi. In P. P. Betancourt and S. C. Ferrence (eds) Metallurgy: Understanding How, Learning Why. Studies in Honor of James D. Muhly, 189–202. Prehistory Monographs 29. Philadephia, INSTAP Academic Press.

Hauptmann, A., Maddin B. and Prange, M. (2002) On the structure and composition of copper and tin ingots excavated from the shipwreck of Uluburun. Bulletin of the American Schools of Oriental Research 328, 1–30.

Lo Schiavo, F., Stech, T., Maddin, R. and Muhly, J. D. (1987) Nuragic metallurgy in Sardinia; second preliminary report. In S. M. Balmuth (ed.) Studies in Sardinian Archaeology III. Nuragic Sardinia and the Mediterranean World, 179–187. British Archaeological Reports International Series 387. Oxford, British Archaeological Reports.

Lo Schiavo, F., Giumlia-Mair, A., Sanna U. and Valera, R. (eds) (2005) Archaeometallurgy in Sardinia from the Origins to the Beginning of the Early Iron Age. Instrumentum 30. Montagnac.

Lo Schiavo, F., Muhly, J.D. Maddin R. and Giumlia Mair, A. (eds) (2009) Oxhide Ingots in the Central Mediterranean. Biblioteca di Antichità Cipriote 8. Rome, A. G. Leventis Foundation and CNR Istituto di Studi sulle Civiltà dell’Egeo e del Vicino Oriente.

Maddin, R. (1989) The copper and tin ingots from the Kaş shipwreck. In A. Hauptmann, E. Pernicka and G. A. Wagner (eds) Old World Archaeometallurgy, Der Anschnitt, Beiheft 7, 99–105. Bochum, Deutches Bergbau-Museum.

Maddin, R., Muhly J. D. and Stech Wheeler, T. (1983), Metal working. In T. Dothan and A. Ben-Tor, Excavations at Athienou, Cyprus 1971–1972, 132–138. Qedem 16. Jerusalem, The Hebrew University.

Muhly, J. D. (1973) Copper and Tin. The Nature of the Metals Trade in the Bronze Age. Hamden, CT. (Transactions of the Connecticut Academy. of Arts and Sciences, vol. 43).

Muhly, J. D. and Maddin, R. (1988) Report on analysis of fragment of copper oxhide ingot (no189) from Maa Palaeokastro. In V. Karageorghis and M. Demas Excavations at Maa-Palaeokastro 1979–1986, 471–473. Nicosia, Department of Antiquities, Cyprus.

Muhly, J. D, Maddin, R. and Karageorghis, V. (eds) (1982) Early Metallurgy in Cyprus, 4000–500 BC. Nicosia, Pierides Foundation.

Muhly, J. D., Maddin, R. and Stech, T. (1988) Cyprus, Crete and Sardinia: Copper oxhide ingots and Bronze Age metals trade. Report of the Department of Antiquities, Cyprus, 281–298.

Muhly, J. D., Maddin, R. and Wheeler, T. S. (1980) The oxhide ingots from Enkomi and Mathiati and Late Bronze Age copper smelting in Cyprus. Report of the Department of Antiquities, Cyprus, 84–99.

Muhly, J. D., Wheeler, T. S. and Maddin, R. (1975) Ingots and the Bronze Age copper trade in the Mediterranean: a progress report. Expedition 17(4), 31–39.

Stech, T., Maddin, R. and Muhly, J. D. (1985) Production at Kition in the Late Bronze Age. In V. Karageorghis and M. Demas (eds) Excavations at Kition V: The Pre-Phoenician Levels, Areas I and II. Part I, 388–402. Nicosia, Department of Antiquities.

Tholander, E. (1971) Evidence for the use of carburized steel and quench hardening in Late Bronze Age Cyprus. Opuscula Atheniensia 10, 15–22.

2. Late Bronze Age copper production in Cyprus from a mining geologist’s perspective

George Constantinou

---

Introduction

The first metals that attracted the attention of Neolithic man because of their bright shiny colours were native gold and copper which were then considered as other types of rocks. People realized that these nuggets of gold and copper were much heavier than all the other stones, but were more surprised when they tried to break them. They changed shape with hammering and the gold could be used for jewels. The copper, with hammering, in addition to changing shape was also hardened and thus was ideal for making blades, hooks, arrow heads and spear heads as well as other useful objects which were previously made with great difficulty from small hard stones. Later it was discovered that with sufficient heating in a fire these copper objects softened once again. Further hammering yielded metal for making more complex and much finer artefacts. A further development was melting and casting in moulds, which greatly increased the variety and quality of the copper objects.

The occurrence of native copper in many locations within southern Europe and the eastern Mediterranean was associated with oxide, silicate and carbonate minerals of copper, the impressive colours of which attracted the ancients, who with their experience in the melting of copper succeeded in extracting copper metal from these ores, thereby increasing considerably the availability of the metal. Copper oxides occur in all countries of the Eastern Mediterranean, mostly as coatings on host rocks, and because of their colours are easy to discover and exploit, but the amount of copper they can produce is very limited. In Yugoslavia (Jovanovic 1980) hammered native copper was produced c. 5000 BC and smelted copper by 4300 BC, and in northern Greece (Renfrew 1973) copper objects first appear c. 4800 BC. In Çayönü Tepesi, SE Anatolia, hammered native copper was used c. 7000 BC, whereas smelted copper was used much later (Muhly 1989). The total number of copper objects found in Yugoslavia is more than in Anatolia, despite the latter having more voluminous copper deposits. These deposits, however, are in the form of sulphide ores, the smelting of which to produce copper was not then known.

A characteristic of this early period is that although copper was a most sought-after metal, and its production from the smelting of its oxide ores was practiced in many areas of the Eastern Mediterranean, the quantity produced hardly satisfied the demand in the area of production, and there is no known centre of large-scale production or a route of transport to areas of consumption. In contrast, in the Neolithic period the most sought stone was obsidian and already in the 7th millennium BC three source areas (Lake Van, Cappadocia, island of Melos) had been recognized, and artefacts from these have been found in settlements as far as 3000km away.

An exception is the copper in ornaments from Nahal Mishmar, found in a cave in the Judaean Desert in Israel. The high arsenic content (1.5%–11%) of the copper-arsenic alloy of the ornaments is notable, and the recent geological literature suggests the Caucasus as its source. This is interpreted as a long distance cultural contact. Its production from ore rich in enargite (Cu3AsS4) and tennantite (Cu8As3S7) minerals which were roasted before smelting (Key 1980) indicates that copper extraction from sulphides was known at the time in the Caucasus. However, the production of copper from this area was limited by the small size, vein type ore bodies, but also mainly because of health hazards from the noxious gases of arsenic from the roasting and smelting of these ores. It is evident that although the copper metallurgy and metalworking in the eastern Mediterranean was very advanced in the 4th millennium BC, production could not satisfy the demand. This Chalcolithic period lasted up to the second half of the 3rd millennium BC.

Cyprus was blessed by nature with many large copper deposits, but in the Chalcolithic period the appearance of copper took place later than in the surrounding countries, its use was limited and the copper metalworking relatively primitive. The first tools were made of hammered native copper and later on of smelted copper. All of them have been found in the SW part of the island. They include a hook dated to c. 3500 BC (Peltenburg 1982) a chisel of c. 3200 BC (Dikaios 1936) and one blade, one axe and one adze of c. 2500 BC (Peltenburg 1982). The probable sources of the copper used were small occurrences of coatings of oxides, silicates and carbonates of copper on the allochthonous pillow lavas of the Mammonia Complex which were thrust on the autochthonous rocks of the area during the collision of the African and Eurasian Plates.

Mining geology and the beginning of the exploitation of the copper sulphide deposits of the Troodos Ophiolite Complex in antiquity

The Troodos Ophiolite Complex was formed by submarine volcanic activity in a deep ocean called Tethys, 90 million years ago. The submarine volcanic activity was associated with the formation of many sulphide copper deposits which made Troodos and Cyprus one of the five richest countries in copper in the world per unit area. These copper deposits after their formation were covered with younger volcanic and sedimentary rocks 3000m thick. Under this thick cover their discovery and exploitation is not feasible even today. A series of unique and complicated geological processes which lasted millions of years uplifted the Troodos Ophiolite to its present impressive elevation of 1951m above sea level and gave birth to an island which in the course of its history was named Cyprus. The impressive topography affected the island’s climatic conditions, considerably increased the annual precipitation and thus the water resources, and formed through weathering a thick cover of fertile soil. Thus conditions were established that favoured agriculture and the establishment of a thick forest cover.

A consequence of the rapid differential uplifting of Troodos was extreme erosion that removed not only the sediment cover but also exposed on the surface all the island’s constituent geological rock units including its volcanic and sulphide copper deposits. The exposed sulphide deposits underwent extensive subaerial oxidation which was promoted by a series of favourable geological and climatological conditions acting separately or in synergy. The increase of the annual precipitation, particularly its distribution through the twelve months of the year was a key favourable factor. The meteorological data indicate that 75% of the annual precipitation occurs in the three months of winter and 10% in November and March. This precipitation considerably exceeds the total evapotranspiration of the five months, so the excess rainwater penetrates into the deeper rocks including the sulphide deposits. The deep penetration of the rainwater is further favoured by the intense fragmentation of the rocks, a result of the uplifting of Troodos. The high porosity of the Cyprus sulphide ores further increased the penetration of the rainwater.

The primary chemical composition of the Troodos sulphide deposits is very simple. They mostly consist of iron and sulphur, small and variable amounts of copper (0.5%–4.5%) and minor zinc. The primary mineral composition is also simple consisting mostly of iron pyrite (FeS2), small and variable amounts of chalcopyrite (CuFeS2) and minor sphalerite (ZnS). Their chemical and mineralogical compositions are interpreted to reflect the submarine environment of their formation at the axis of seafloor spreading, and in the recent geological literature are classified as Cyprus Type Deposits.

During the oxidation process the primary minerals react with the rain water and the dissolved oxygen in it, forming an acid solution which contains various amounts of sulphuric acid, ferrous sulphate, copper sulphate and minor amounts of zinc sulphate. The chemical reactions are as follows:

iron pyrite: FeS2+ H2O + 3.5 O2→ FeSO4 + H2SO4;

chalcopyrite: CuFeS2 + H2O + 3.5 O2→ CuSO4 + H2SO4;

sphalerite: ZnS + H2O+ 3.5 O2→ZnSO4+H2SO4.

The presence in the acid solution within the Cyprus deposits of the autotrophous bacteria Theobacillus ferroxidant, which get the energy they need from the oxidation of sulphur and iron, promotes the formation of the strong solvent ferric sulphate (Fe2(SO4)3) in the acid solution and further increases the oxidation of the sulphide ore.

A small part of the acid solution is lost by surface draining but the major part penetrates into the copper deposit. On the way the iron sulphates of the acid solution hydrolyze and precipitate iron hydroxides. Some of the copper sulphate on its way down precipitates a great variety of copper oxides (Table 2.1). The copper sulphate in solution at or close to the surface reacts with pine resin of the surface soil, is reduced to native copper, and forms thin coatings of malachite on fragments of limestone in the small draining streams. Where the solution fills surface shallow depressions and evaporates in the dry months it precipitates coloured hydrated salts of iron and copper sulphates (Fig. 2.1a). The majority of the acid solution penetrates into the ore bodies and when it reaches the water table the copper sulphate is reduced and precipitates a variety of secondary copper sulphide minerals (Table 2.1) (Constantinou 1972).

The final result of these complicated processes is the formation above the ore bodies of an ‘iron hat’ (gossan) the thickness of which varies from several to 40m with bright red and yellow colours (Fig. 2.1b). Under the gossans in the upper part of the ore bodies are the zones of secondary copper enrichment in which the copper content increases from ≥0.5% up to 25%. In the enriched zones in which the dominating secondary copper sulphide is chalcopyrite the copper content is up to 15% (Fig. 2.1d), and where the dominating copper minerals are chalcocite, covellite and bornite (Fig. 2.1f) the copper content is 25% because copper minerals contain higher copper than chalcopyrite (Table 2.1). It is worth mentioning that the exploitation of the Cyprus copper ore bodies from 1920 to 1980 produced 1.3 million tons of copper metal in the form of copper sulphide concentrates. The mined ore, containing copper 0.5%–4.5%, was treated in flotation plants and produced iron pyrite and copper sulphide concentrates with copper content 20%–25% which was exported to Europe for smelting. Nature, through oxidation, solution and reprecipitation, concentrated copper sulphides in the upper part of some ore bodies to the level acceptable to modern furnaces. During the first years of operation of Skouriotissa mine in 1920 the mined ore was exported to furnaces in Europe without treatment.

Table 2.1. Secondary copper minerals

Fig. 2.1. a) secondary copper sulphates minerals, b) gossan (South Mathiatis mine), c) native copper, d) high grade copper ore (rich in chalcopyrite), e) ancient (8000 BC) water well 12.5m deep (Kissonerga, SW Cyprus), f) high grade copper ore (rich in chalcocite), g) ancient timber support of pine wood with native copper replacing its resin.

The ancient miners were attracted by the bright colours of the gossans and found on the surface small amounts of native copper oxides and copper sulphates. They exploited these, and started exploring the lower parts of the gossans to find more by digging shafts, a technique which was known in south and south-western Cyprus since 8000 BC (where they dug wells for finding groundwater in periods of prolonged drought, to depths of 12.5m (Fig. 2.1e)). In some ore bodies the shafts penetrated the gossans and reached the zones of secondary copper enrichment. If the copper in these zones had been in the form of oxide minerals the exploitation of the Troodos copper deposits would have started earlier than the 4th millennium BC; but it was in the form of copper sulphide minerals and the extraction of copper from their smelting was then not known, being more complicated and involving roasting to convert the sulphides to oxides, and then melting and re-melting to produce copper metal.

Archaeological data indicate that during the Chalcolithic in SW Cyprus, in addition to the experience of extracting copper from smelting its oxide ores, people also had considerable knowledge of pyrotechnology for the production of lime from burning limestone, and for handling fire in pottery kilns to produce selectively red or black fired ceramics. Such experienced craftsmen could roast the copper-rich sulphide ore underlying the gossans, in their effort to smelt it. The copper oxides produced from the roasting gave them the stimulus to continue with new efforts to smelt the ore. When they succeeded, part of the high copper content separated from the produced copper matte and flowed as copper metal. They had the experience to repeat the procedures of their ‘invention’ and to proceed with its refinement.

Their effort for refinement was assisted by some favourable geological factors present in the Cyprus copper mines. The gossans contain a high proportion of amorphous silica as well as iron hydroxides and natrojarosite which constitute a very good flux for lowering the melting point and lowering the viscosity (fluidity) of the melt and for better separation of the copper metal. The umbers that occur above some of the ore bodies or in their vicinity consist of an amorphous mixture of iron with manganese hydroxides and small amounts of amorphous silica, making them very good fluxes. Furthermore the shallow depth of the ore bodies and the conglomeratic structures of their ore make underground mining easy, even with primitive tools, for the experienced miners of the island. The surrounding thick forests supplied the necessary energy for the pyrometallurgical treatment of the copper ores (Constantinou 1982). All these, but in particular the extraordinary secondary concentration of copper in the upper parts of the deposits, enabled the Cypriot miners to produce for the first time in the Eastern Mediterranean more copper than the local demand and to export the surplus to the surrounding countries.

In some countries of the Eastern Mediterranean and the Near East there are large copper sulphide deposits. The largest copper deposit in Turkey is the Ergani Maden area in SE Anatolia. The ore body is classified as Cyprus Type Deposit (Saltiray et al. 1976) because its geological setting, chemistry, mineralogy and submarine environment of formation are very similar to the Troodos copper deposits. The average copper content of this deposit is 1–2%. Part of the ore body was exposed on the surface, oxidized and formed gossans but the prevailing conditions did not produce zones of secondary enrichment with high copper content. Thus the production of copper at that time was not possible, or limited, and so the Hittites imported copper from Cyprus in the 15th century BC (Goetze 1959). The second largest massive sulphide copper deposit of Turkey is Kure near Samsun, close to the Black Sea coast. The ore of the deposit has similarities with the ore of the Troodos deposit but its average copper content is ~1%. In the Near East there are the massive sulphide deposits of the Oman Ophiolite which are classified as Cyprus Type Deposits. Their copper content varies from 1% to 2%. Babylon and Mari were importing copper from Cyprus at the beginning of the 18th century BC (Muhly 1972).

In the eastern Peloponnese in Greece is the large cupriferous pyrite deposit of Ermioni, close to Tiryns and Mycenae, which contains up to 3% copper (Marinos 1982). Part of it was exposed on the surface, oxidized and formed a gossan but the secondary enrichment contained less than 5% copper. Its exploitation at that time was difficult, and Cyprus supplied copper to Minoan Crete in the 16th century BC, and later to Mycenae and mainland Greece. In Saudi Arabia there is the large massive sulphide deposit of Jebel Said containing 1% to 2% copper associated with lead and zinc sulphides, but despite this potential source Cyprus was exporting copper to Egypt in the early 15th century BC (Georgiou 1979). The mining geology data suggest that Cyprus, because of its unique geology, was the dominating producer and exporter of copper in the 2nd millennium BC in the eastern Mediterranean and the Near East. All of the above explains the expansion of the production of copper in Cyprus in the Late Bronze Age, with a Cypriot location for Alashiya; any contrary suggestions point to a lack of knowledge about the mining geology of this region.

The production of copper in Cyprus was intensified in the Late Bronze Age (1600–1100 BC) because of the considerable increase in its use and demand in the Eastern Mediterranean. The main reason for that was the sudden easy availability of tin in the area. The alloy of copper with tin at a ratio 10:1 has much better physical and mechanical properties than pure copper for making weapons, tools and household utensils. The addition of tin in copper also improves the motility of the melt for better casting and can produce finer bronze objects. The dominating supplier for copper was Cyprus but the source of the tin has been debated for a long time in the archaeological and archaeometallurgical literature.

The Caucasus (Field and Prostov 1938), Afghanistan (Maddin et al. 1977; Muhly 1985) and Turkey (Yener et al. 1989) have been suggested as sources of tin for this period, but none was generally accepted and Afghanistan remained as the most probable source. Ancient texts of this period report that there was a radical decline of the price of tin in Nuzi, and Heltzer (1978, 109–111) suggested the probable existence of a tin source in its area. Recently J. Phillipson and C. Lambrou Phillipson (forthcoming) based on geological, mining and archaeological data and ancient texts of this period proposed the area of Nuzi in Mesopotamia as the source of Late Bronze Age tin. They suggested that the radical change in the availability of tin and the slump of its known price in silver equivalent from 6:1 in the Middle Bronze Age to 520:1 during the Late Bronze Age was due to the discovery of placer tin deposits in the area of Nuzi which to a mining geologist is very plausible.

Table 2.2. Minerals of economic importance recovered from placer deposits.

Fig. 2.2. (Colour Plate 1) Geological map of Middle East 1:5.000.000 (Geological Survey of Iran 1986).

Placer (alluvial) deposits are surficial concentrations of economic mineral particles deposited by flowing water or air. Placer gold mining is the most ancient kind of mining and very often the gold particles were associated with cassiterite (tin) particles. At present placer deposits supply 80% of the world production of tin and considerable part of the world production of the economic minerals listed in Table 2.2. The source of the economic minerals of placer deposits can be hundreds of kilometres away, and is the product of weathering of huge volumes of source rocks over a long period of geological time. They can be of any geological age but most are geologically young. However a special group of Palaeoplacers is the gold deposits of Witwatersrand in South Africa, about 2.7 billion years old. A mineral, to form placer deposits, must be resistant to chemical decomposition, hard or malleable to resist fragmentation during its transport for long distances and more dense (≥3.3 g/m ³) than the common forming minerals (Table 2.2).

The sources of cassiterite for tin placers are granitic rocks especially greisens, an acid igneous rock which often occurs in the form of veins intersecting granite cupolas and surrounding rocks. Greisens consist of quartz, mica, topaz and small amounts of cassiterite. Nuzi is very near a river draining the western flanks of the Zagros mountain chain. In that area there are many granitic bodies, indicated by arrows in Fig. 2.2, which range in size from batholiths to stocks with greisens veins intersecting some of them. Weathering of granitic bodies and greisens will decompose the soft mica, and the other mineral particles of the weathered rock debris will be washed by the rivers draining the area. The flowing water will concentrate the dense cassiterite particles where the morphology of the river bed forms natural traps. The underground geology may influence the formation of these traps but has no genetic relation with the cassiterite.

Table 2.3. Raw materials for core-formed opaque glass vessels.

The discovery of a tin placer by the ancient prospectors was easy because the black colour of cassiterite is different from the light colour of the associated quartz and topaz. Its exploitation was simple, demanding only running water, primitive tools and experience of thousands of years from earlier exploitation of gold placers. No shafts and galleries were needed, no specialized miners, and thus the cost of labour was low. Because of the high content (80%) of tin in cassiterite fuel demands would be limited and thus no forest for smelting on the site was necessary. Finally and for the same reason no slag heaps are expected to occur. The cost of production of placer tin is very low and until the sixties of last century in Bolivia a tin placer containing 100ppm tin was exploitable economically, whereas for primary hard rock tin deposits the economic grade was a minimum 10000 ppm of tin (UNIDO 1969). This coincides with the order of magnitude reported for the difference between the price of tin in the earlier Middle Bronze Age Kanesh and in Old Assyria and at Nuzi in the 15th Century BC. The easy and uncontrolled exploitation by individuals of various tin placers made the supply of the metal much higher than its demand and brought about the radical decline of its price.

The archaeological literature indicates that Nuzi in the 15th century BC had the biggest collection of core-formed opaque glass vessels and was an important glass working centre in the Near East. Other centres of similar glass workings were at Assur and Tell Brak, all beginning in the early 15th century BC, contemporary with the Kassite period of Babylonia whereas glass is rare elsewhere at this period (Oppenheim et al. 1970). All of these areas are located near rivers draining the western flank of the Zagros Mountains.

The mining geology explanation suggested by Phillipson and Lambrou Phillipson (2004) is that the boom in the local glass industry was due to the discovery and exploitation of placer tin. The by-products of the exploitation of the placer tin were quartz sand, topaz and fine cassiterite, the raw materials required for the production of core-formed opaque glass (Table 2.3). Quartz supplies the silica of the glass, topaz the alumina, and fluorine which is a flux, and unrecoverable very fine cassiterite the opacifier. All these also suggest that the tin production in Mesopotamia was in the hands of Kassites, and very rightly the Greeks who were important importers of tin at that time named the product of the Kassites ‘kassiteros’.

The radical change in the availability of tin and the slump in its price dramatically favoured the Cypriot producers of copper because the exchange of copper was bringing in much more tin compared to the equivalent in the Middle Bronze Age. As a result the Cypriot producers of copper accumulated much more tin than they needed in the island and they had to export the surplus. So part of the trade of tin at that time was in the hands of Cypriots who exported the tin surplus to Egypt and further west (Kassianidou 2003). This further increased the export of copper from Cyprus because a cargo of copper and tin was more attractive than freight consisting only of pure copper.

A convincing example is the early 14th century BC shipwreck of Uluburun which was carrying about ten tons of copper and one ton of tin (Bass 1985), the right ratio for bronze. The expansion of export of copper to Egypt during the Amarna period in the 14th century BC is testified in the seven letters exchanged between the King of Alashiya and the King of Egypt and one letter exchanged between the Governor of Alashiya and the Governor of Egypt. They refer to a total of 897 copper ingots of a total weight of 27 tons of copper exported to Egypt in exchange for various valuable commodities (Knapp 1985, 1996; Muhly 1972). Exports of Cyprus copper extend in this period to east and west as well as to north and south (Fig. 2.3).

The geographical position of the island at the crossroads of three continents, Asia, Africa and Europe, in the middle of an area where the great civilizations of this period flourished, made Cyprus a centre of maritime trade and cultural contacts. The island was supplying the surrounding countries with the basic material for their technological development in exchange for valuable products, thus considerably raising the standard of living of its inhabitants as is witnessed by the valuable material objects found in the tombs of this period. In this period many mining and metallurgical centres for the exploitation of the copper deposits and the primary production of copper were operating in the periphery of the Troodos mountain range (Fig. 2.4). The metal was transported overland to the coastal cities for further refinement, smelting and casting in the abundant metal works. The coastal cities had monumental buildings with building stones that indicate developed techniques of their quarrying, loading, transporting, dressing and laying.

Fig. 2.3. Map showing the routes of copper and tin trade.

The high value and weight of copper, tin and the other commodities exchanged and their transport by sea over long distances made necessary the development of shipbuilding technology for larger and safer ships. The island became an important shipbuilding centre in antiquity when, as Strabo remarked 1500 years later, ‘… men sailed over the sea without fear and with large fleets’. In the coastal cities suitable harbours were constructed for easy and safe loading and unloading of the valuable commodities of the maritime trade. The cities and their harbours were located in coastal inlets that offered protection from the winds and for thousands of years facilitated contacts by sea with the neighbouring countries.

These inlets were formed during the final configuration of the island as a result of the interplay between uplift and sea level fluctuations during the Pleistocene glaciations. During glacial periods much seawater is held in the ice sheets and the sea level may drop by more than 120m. The coastline extended further out to sea by as much as several kilometres and the surface area of the island was 25% bigger than at present. The extra land was subject to erosion and crossed by significant valleys. Some of the first permanent settlements of the island were probably situated in this extra land, when the sea level was 30m lower. When the ice sheets melted sea level gradually rose to its present level, drowning the lowermost reaches of many valleys forming the above-mentioned important inlets. It also drowned the early permanent settlements and their cultivated land. The fear of a new ‘cataclysm’ forced the location of the settlements of the island from Neolithic to Middle Bronze Age to be on safe elevations away from the sea.

The introduction of the oxhide ingots in Cyprus

The Late Bronze Age in Cyprus is characterized by the appearance of copper ingots with an oxhide shape. The cargo of the Uluburun shipwreck included the largest collection of copper ingots ever found. Some of the copper ingots were of discoid and elliptical shapes but the majority of them had the shape of a dry oxhide (Pulak 2000) which appeared in the 16th century BC (Muhly et al. 1988) and disappeared in the 11th century (Lo Schiavo 1998). This shape is an improvement on a former rectangular shape, and is concave on all sides with a protrusion at each corner (Fig. 2.5). The improved shape facilitated its handling and transport (as depicted in ancient Egyptian tombs), and storage in the hold of ships as well as in storerooms. It proved suitable for other metals such as tin and lead not produced in Cyprus but traded by Cypriots in different sizes, as shown in storerooms in Egypt with a note of the type of metal (Buchholz 1959).

Fig. 2.4. (Colour Plate 2) Mineral resources map of Cyprus with locations of ancient copper mines and slag heaps.

Fig. 2.5. Miniature copper oxhide ingot from Enkomi in Cyprus, with Cypriot marking system.

The weight of the Cypriot oxhide copper ingots varies from 20 to 29kg but the majority of them are about 25kg. The 360 oxhide copper ingots of the Uluburun shipwreck have different weights, most probably due to the fact that they were cast in moulds made with sand rather than stone, as indicated from the lack of any of the damage that would be expected from the use of a metallic lever to remove them from the stone mould (Buchholz 1959).

Archaeological excavations suggest that the production of copper in Cyprus in this period was related to religious institutions. This is indicated by the copper statuettes standing on copper bases with oxhide shape: the male ‘Ingot God’ from Enkomi and the other a female, the ‘Bomford’ Goddess of unknown origin. In sanctuaries of this period were found models of copper oxhide ingots with signs in Cypro-Minoan script. In some cases these are associated with ships’ anchors, demonstrating the intimate connection of the island with copper and the sea. This shape of copper ingot is depicted on seals and in Egyptian tombs of this period. They are also shown as part of the imagery on later four-sided copper stands.

In this period in Cyprus the still undeciphered Cypro-Minoan script, with elements of the Linear A, was introduced. Also introduced at this time was the Cypriot marking system for the copper and tin ingots. These also coincide with the beginning of the export of Cyprus copper to Minoan Crete and mainland Greece. The majority of the copper oxhide ingots are marked with symbols many of which are identical with signs of the Cypro-Minoan script, either stamped at the final stage of casting or incised after the solidification of the metal. The Cypriot system of marking was found on disc- and oxhide-shaped ingots of tin from the Uluburun shipwreck and on the tin and lead ingots from the Khar Samir shipwreck on the coast of Haifa (Galili et al. 1986). This suggests that the seaborne trade in these metals was controlled by the Cypriots (Kassianidou 2003). In addition to the mining geology, metallurgical, archaeological and lead isotope data the Cypriot system of marking the oxhide ingots testifies that their copper was produced in the copper mines of Cyprus.

The appearance of the Cypriot copper oxhide ingots coincides with the discovery of placer cassiterite and the dramatic increase in the availability of tin in Nuzi, Mesopotamia, and its disappearance with the dearth of tin metal in the 12th century BC. The high tin content of cassiterite (80%) made possible its direct exchange as tin metal. The grains of placer cassiterite are small, easy to measure by weight or volume, easy to handle and recognize, are durable and of high value and became a medium of exchange. There was an Akkadian expression ana quatin (tin of the hand) in the Kanesh tablets, for a commodity which was used to pay taxes and as a money substitute, and it has been suggested by Phillipson and Lambrou Phillipson (2004) that this was placer cassiterite. It is also known from texts that ‘tin of the hand’ became a medium of exchange in the Late Bronze Age (Oppenheim et al.1970).

On the other hand, in the Late Bronze Age the easy availability of placer cassiterite (tin) stimulated the expansion of production and export of copper in Cyprus. Most of the exported copper was used with tin in the metal workshops of the surrounding countries for making bronze. In this period copper and tin were the most sought metals and they were also exchanged with the other known metals gold and silver and other valuable commodities. However the greater part of the copper of Cyprus was exchanged with cassiterite (tin) which had been established much earlier as a medium of exchange. As mentioned before, the first reported exchange of copper of Cyprus was with Mari (1982–1739 BC) and Babylon (1750–1725 BC) when in the latter the functions of its society were regulated by the Laws of Hammurabi (1792–1750 BC). The oxhide ingot shape of copper was probably introduced as standard of the copper of Cyprus for its exchange with the cassiterite of Mesopotamia. The incremental flexibility of cassiterite grains, small amounts of which could easily be measured, was easily adjustable to the small deviations of the weight of the oxhide ingots. The Cypriot marking system was probably introduced for copper and tin oxide ingots exchanged with Minoan Crete, mainland Greece, and probably Egypt. In modern times, standardization certification, quality control and marking were introduced first in countries with advanced technology and high cultural levels. The oxhide copper ingot of Cyprus and the Cypriot system of marking most probably was a Late Bronze Age standard with marking, introduced by people with advanced technology, highly organized trade and high cultural level.

Acknowledgements

I wish to thank the organizers of this symposium Vasiliki Kassianidou and George Papasavvas for inviting me to participate. Thanks are also due to my colleagues Costas Xenophontos and Ioannis Panayides for fruitful discussions during the preparation of this paper. Last but not least, I am indebted to Stalo Constantinou for her assistance in typing the manuscript.

Bibliography

Bar-Adon, P. (1980) The Cave of the Treasure. Jerusalem, Israel Exploration Society.

Bass, G. F (1986) A Bronze Age shipwreck at Ulu Burun (Kaş), 1984 campaign. American Journal of Archaeology 96, 269–296.

Buchholz, H. G. (1959) Keftiubarren und Erzhandel in zweiten vorchristlichen Jahrtausend. Prähistorische Zeitschrift 37, 1– 40.

Constantinou, G. (1972) The Geology and Genesis of the Sulphide Ores of Cyprus (unpublished Ph.D. thesis). Imperial College, University of London.

Constantinou, G. (1982) Geological features and ancient exploitation of the cupriferous sulphide orebodies of Cyprus. In J. D. Muhly, R. Maddin and V. Karageorghis (eds) Early Metallurgy in Cyprus 4000–500 BC, 13–24. Nicosia, Pierides Foundation.

Dikaios, P. (1936) Excavations at Erimi 1933–1935: final report. Report of the Department of Antiquities Cyprus, 1–81.

Field, H. and Prostov, E. (1938) Tin deposits in Caucasus. Antiquity 12, 341–345.

Galili, E., Shmueli, N. and Artzy, M. (1986) Bronze Age ship’s cargo of copper and tin. International Journal of Nautical Archaeology and Underwater Exploration 15/1, 27–37.

Georgiou, H. (1979) Relations between Cyprus and the Near East. Levant 11, 84–100.

Goetze, A. (1956) The inventory I Bo T131. Journal of Cuneiform Studies 10, 32–38.

Heltzer, M. (1978) Goods, Prices and the Organization of Trade in Ugarit. Marketing and Transportation in the Eastern Mediterranean in the second half of the II Millennium B.C.E. Wiesbaden, Ludwig Eichert Verlag.

Jovanovic, B. (1980) The origins of copper mining in Europe. Scientific American 242(5), 152–167.

Kassianidou, V. (2003) The trade of tin and the island of copper. In A. Giumlia-Mair and F. Lo Schiavo (eds) The Problem of Early Tin. Acts of the XIVth UISPP Congress. British Archaeological Reports, International Series 1199, 109–119. Oxford, Archaeopress.

Key, C. A. (1980) The trace element composition of the copper alloy artefacts in the Nahl Mishmar hoard. In Bar-Adon 1980, 238–243.

Knapp, A. B. (1985) Alasiya, Caphtor/Keftiu, and eastern Mediterranean trade: recent studies in Cypriote archaeology and history. Journal of Field Archaeology 12, 231–250.

Knapp, A. B. (1996) Introduction. In A. B. Knapp (ed.) Near Eastern and Aegean Texts from the Third to the First Millennium BC. Sources for the History of Cyprus Vol. II (eds P. W. Wallace and A. G. Orphanides). Altamont, New York, Greece and Cyprus Research Center.

Lo Schiavo, F. (1998) Sardinian oxhide ingots. In T. Rehren, A. Hauptmann and J. D. Muhly (eds)