Sunday, 26 January 2020

Emily Loughton Dissertationelfig10

 1.0 Introduction

1.1 Introduction

In 1991 a group of divers from the South West Maritime Archaeological Group (SWMAG) discovered a hoard of forty-four tin ingots at the mouth of the River Erme, in south west Devon. Tin ingots were the only artefact recovered from the site due to the nature of the investigation which was primarily based on the use of an underwater metal detector. The discovery has led to a wide discussion that is primarily concerned with the age of the ingots, with many papers talking of them being British Bronze Age in date (McDonald 1993, 27); however this assumption does not appear to have been based on much substantial evidence as relatively little work has been carried out on the ingots and the use of tin in the Bronze Age in general.

This paper hopes to explore this assumption further by examining the available evidence and attempting to establish how likely it is that the tin ingots are British Bronze Age in date.

1.2 Background

This project has been started almost from scratch due to the lack of work that has been carried out on this particular site, and although in some ways this has been beneficial to the author, as it has allowed a relatively wide collection of material to be brought together in order to narrow down the search for evidence relating to the question, it has also had its problems. With forty-four tin ingots, and nothing else to go by it was difficult at first to know where to start. The question is related to the Bronze Age because this date has been discussed in the past and since it is the earliest date mentioned it seems necessary to start there and study the available evidence to see how the ingots could potentially fit into this period.

The Bronze Age in general is a well publicised period, however when it comes to researching maritime features such as trade and prehistoric vessels this often presents a problem. It should be noted that this problem is not only concerned with the Bronze Age but is found throughout the ages due to the slow integration of maritime archaeology into the wider discipline of terrestrial archaeology. Muckelroy, as a forerunner of maritime archaeology in Britain has proven to be the most useful author in this sense and his papers on prehistoric cross-channel exchange (Muckelroy 1981) and work on two of the most promising Bronze Age maritime sites in southern England, Langdon Bay and Moor Sands, along with Baker (Muckelroy & Baker 1979:1980) has helped a great deal in writing this paper. McGrail (1998:2004) has provided substantial evidence for vessels of the period which has also proven extremely useful.

The study of ancient tin has received little attention in the published literature over the past years. Many pieces of literature concerned with archaeometallurgy have focused on more prominent metals such as copper or bronze with little thought given to tin, even though it is an essential ingredient. Another problem to have arisen from this research is that many of the relevant sources that do comprise information on tin were written a number of years ago (Thibault 1908) and so it is evident that this is a discipline in desperate need of being updated. Tylecote (1987:1990) and Penhallurick (1986) have both provided a great insight into the extraction, smelting and casting of tin in the early periods.

Fleming (1978:1988) has been the most useful author when examining the land use of Dartmoor in the Bronze Age and along with Fox’s paper on the Avon Valley (Fox 1957) this has allowed much to be learnt. Price (1993) has provided a wonderful insight into the settlement pattern on Dartmoor in relation to streamworks, a subject that has had little discussion amongst other authors.

Bass has written some excellent papers on both the Cape Gelidonya and the Ulu Burun shipwrecks which have proven to be invaluable to this paper due to the nature of finds which included large quantities of copper and tin ingots on both sites.

1.3 Aims and objectives

Aim: The overall aim of this paper is to establish the likelihood that the tin ingots from the Erme Ingot Site are British Bronze Age in date and to demonstrate how they could have fitted into Bronze Age activities such as the production and trade of tin.

Objectives: In order to carry out this aim effectively a number of objectives needed to be met and thus appear throughout this paper:

  • A study of the prehistoric      production of tin in Britain and its importance to Bronze Age communities      has been made to see how the ingots could have fitted in to this.
  • X-ray analysis was carried      out in order to gain a further understanding of the composition of the      ingots.
  • A macroscopic examination of      the ingots was made and they have been compared with known Bronze Age      examples from other sites to see how they contrast.
  • Trade routes of this period      have been looked at to identify how these particular ingots may have      fitted into the trade and exchange of such materials around the South West      and beyond this.
  • A discussion of known wreck      sites comparable to that of the Erme Ingot Site and ritual deposits has      been made in order to establish the likely origin of this site.
  • Dartmoor has been examined      for its well known source of tin and the relationship between Bronze Age      settlements and streamworks has been identified in the attempt to      establish the likelihood that Bronze Age communities were extracting and      working metal in this region. The relationship between Dartmoor and the      Erme estuary ingot site has also been investigated.

1.4 Methodology

In order to examine the likelihood that the tin ingots from the Erme estuary ingot site are of British Bronze Age date it was essential to research the factors mentioned in the objectives above. Due to the nature of the project, the investigations were mainly desk based however some practical work was carried out in order to further the author’s knowledge of the composition of the tin ingots. Desk based research was conducted via visits to local libraries and the University library as well as carrying out internet research. From this work a large quantity of material was gathered and read through, and key points were identified.

Practical work included a visit to the Royal Albert Museum in Exeter to view the ingots and gain a greater insight into their composition and form. Permission was given on the day to take six of the ingots back to Bournemouth University to carry out some further analytical work on them.

This included x-ray analysis which was carried out in the attempt to learn more about their structure and form.

The small amount of practical work carried out for this paper has highlighted that fact that further research is needed in order to fully understand the nature of the tin ingots and establish a more accurate date for them.

2.0 Prehistoric Tin Mining and Production of Tin Ingots

2.1 Metals and the Bronze Age

Many pieces of literature concerned with the Bronze Age focus on religious and ritualistic aspects of the period, believing that these were dominant factors which influenced the day to day lives of Bronze Age societies. Therefore when examining the ability to turn rock into liquid (and then metal), which even today is considered quite remarkable by some, it is widely assumed that Bronze Age people would have seen it as a magical and wonderful creation that should be treated with due respect and admiration.

In modern day societies a practical implement such as a fork is seen only for the sole purpose for which it was made, however people living in ancient times, especially those in the Bronze Age often viewed such implements, for example a hammer stone, with far more depth and symbolic meaning that go beyond their practical use.

2.1.1   The Importance of Tin to Ancient Communities

Before it was discovered that tin could be added to copper to make bronze, arsenic was sometimes used in Eastern Europe. This must have produced an alloy of less sustainability than tin bronze however, as tin soon replaced it (Parker Pearson 1993, 82). If a greater proportion of tin is added to the alloy then a harder bronze is created, it would seem that the primitive metallurgists understood this and as a result artefacts required for various jobs were produced with varying concentrations of tin (Alexander & Street 1989, 1). The introduction of bronze meant that work such as tree felling could be done in half the time it took using a stone axe, and with half the effort. The cut-off grade for tin (lowest grade which can be profitable) is 0.5% (McLeish 1992, 271) and the high quality grade of Cornish tin (c.79%) suggests that it would have been an extremely valuable material, providing a good economy for a society who could then trade it for other desirable commodities.

The use of tin-bronze appears to have developed more rapidly in Britain than adjacent parts of Europe, suggesting that perhaps British tin was added to European copper (Northover 1982, 51-2).

2.2 Bronze Age methods for Tin Extraction in the South West of England

In recent years, due to a ‘renewed interest in investigating the origins of metallurgy and mining in Britain’ (Craddock & Lang 2003, 21) there have been some developments in the study of prehistoric copper mining in Britain in the discovery of a number of substantial mining works. However this has not been the case for tin mining (Barber 2003, 79). To date there has been no direct evidence found for tin mining in Britain during this period, though this does not mean that it did not take place. Some have suggested that the lack of evidence could be a result of the ancient workings being destroyed by more modern mining (Parker Pearson 1993, 83), whereas others such as Penhallurick (1986, 115) suggest that tin was obtained in the Bronze Age from areas such as Dartmoor by streamworking, or the open-cast working of outcrops, which would leave little evidence for future archaeologists to find. Tin streaming is a method which exploits the alluvial deposits of tin (Fig.1) which have eroded from the rocks which contained them and washed into streams by surface waters. Cassiterite (the ore which tin is found in) is an extremely dense material meaning that it is easily separated from other lighter materials which are washed away (Tylecote 1962, 63).

This method of gathering the tin ore Cassiterite is relatively simple and leaves very little trace of ever having taken place (Craddock & Lang 2003, 176).


Figure 1. The deposition and concentration of alluvial tin deposits which could then be exploited (


In addition to streamworking, tin may have been obtained by digging under the ground and physically separating the ore from the rock using appropriate tools. A written account from the re-opening of the old Twlly mwyn mine near Darren, Talybont, mid-Wales in 1742 gives an indication of the type of people who may have previously exploited the mine, ‘This mine seems to have been wrought in the beginning of time, and before the use of iron was found out, and when mankind knew the use of no tools but stones.’. And further goes on to accurately describe how the ore came separated from the rock (Fig.2), ‘Their methods seem to be this. They made a great fire of wood in the bottom of their rakes…. And when the rock was sufficiently hot they cast water upon it, which shiver’d it; and then with stone wedges, which they drove in with other stones, they work’d their way through the hardest rocks’ (Craddock & Lang 2003, 21).




Figure 2. Prehistoric Copper Mining using stone tools and firing to separate the ore from the rock using thermal-shock techniques (O’Brien 1996, 21 & 23).


Although this is a copper mine in question, it suggests a way in which tin could have been mined in prehistory, however this type of mining leaves much more evidence than the collection of alluvial deposits. As there is substantially more evidence for copper mining in this way than tin mining, it would suggest that tin was collected differently.


The rarity of evidence for tin streaming and the collection of deposits in shallow gravel beds has been demonstrated by Babu’s (2003)ethnographic case study of the central Indian tribes collect tin from gravel beds where vegetational indicators help to locate tin-rich areas (Sarai trees growing on tin-rich soils produce yellow spots on their leaves). Once located and dug, the tribes take the gravel to water and proceed to sift the material in bamboo pans (which would rarely survive the archaeological record), the heavy Cassiterite remains in the basket as the lighter materials are washed away (Fig.3) and afterwards little trace is left behind (Babu 2003, 178).



Figure 3. Indian tribes pan for Cassiterite using bamboo baskets. Little trace is left behind (Babu 2003, 178).


2.3 Smelting and Casting tin in the Bronze Age


Although as yet no direct evidence has been found for smelting metals in the Bronze Age there is some substantial indirect evidence for the working of tin in the form of pieces of tin slag, tin ingots, and small artefacts which indicate that tin was being collected from Britain’s raw materials and worked from within our shores. Slag has been recovered on a few sites such as Dean Moor in the Avon Valley (Fox 1957, 30), Caerloggas near St. Austell and others (Tylecote 1990, 43). Cassiterite pebbles (Fig.4) have also been found on Bronze Age sites which suggest evidence for extracting the ore.


Figure 4. Cassiterite pebbles (Tylecote 1990, 43).


2.3.1 Smelting Processes


Tin ore is smelted at relatively high temperatures (1000ºC) and so would require the use of some kind of furnace. Archaeological evidence consisting of holes in the ground containing ash and sometimes clay (Tylecote 1987, 106) suggests that prehistoric communities may have used small bowl furnaces (Fig.5) to smelt metals. These consist of a hole dug into the ground and clay or stone forming a furnace structure around the pit. Charcoal and Cassiterite would have been put into the structure and burnt, using bellows or prevailing wind to increase the heat. Molten metal would be produced once the right temperatures had been reached (c.1000ºC) and collect in the bottom of the pit. The ingots of metal produced using this method would have been roughly Plano-convex in shape (Gerrard 2000, 20).


Figure 5. Reconstruction drawing of a small bowl furnace which may have been used in prehistory to produce Plano-convex ingots. (Gerrard 2000, 20).


Once again Babu’s (2003) ethnographic case study provides an idea of how prehistoric communities may have smelted tin ore. The Indian tribes use small clay shaft furnaces with charcoal for fuel and the ore is smelted at lower temperatures of 600ºC (Babu 2003, 178). This is the reported minimum temperature required for smelting tin (Hobbs et al. 2002, 13) and suggests that while modern standards propose that 1000ºC is needed, prehistoric communities may have used lower temperatures.


Once an ore has heat applied to it the impurities contained within the ore become separated due to a number of factors including density, solidity, and solubility (Tylecote 1987, 193). These impurities are then oxidised and form a layer on top of the molten metal which can be skimmed off, leaving a purer metal behind. This slag often contains globules of the metal which can be obtained by crushing. The lack of slag finds on prehistoric settlement sites in western and central Europe may be due to the purity of the Cassiterite mineral, which would not produce much waste deposits (Tylecote 1987, 307).


Experimental work carried out using a furnace adapted from those used in modern day South Africa showed that better quality ore gave tin prills and larger masses of tin than lower grade material. It also demonstrated that Plano-convex ingots can only be produced by re-melting the prills and casting them in a mould (Tylecote 1990, 44). However, if a clay bowl furnace was used the Plano-convex ingots would be a primary product and thus the tin would not need to be cast.


A very pure tin could be obtained in prehistory by smelting in a furnace using charcoal, and carrying out drossing (oxidising impurities when molten and skimming off surface) (Tylecote 1990, 46). Refining the metal by re-heating and drossing was likely to have been carried out if necessary in order to produce a purer quality metal.


If tin was not being exploited in Bronze Age Britain, and instead was being imported to the Isles as some have suggested (Barber 2003, 79) then why have such finds as slag- produced in the process of smelting tin been found at early Bronze Age settlement sites?


2.3.2 Casting


Casting metallic tin in moulds has evolved since prehistoric times. Unlike weaponry and tools, ingots did not require such uniform shapes as they were temporary artefacts made with the intention of melting down at a later date. However we do see a similar use of forms over different periods which suggest either convenience or simplicity in their design. In medieval times stone moulds were commonly used which would produce ingots of better shape and also have the advantage of being re-usable. In earlier periods clay was often used for casting metals and for obvious reasons these were one-use only moulds. In prehistoric times the small furnace mentioned above would have smelted the metal and the shapes produced may have been kept or perhaps the material would have been hammered into the desirable shape (Tylecote 1987, 107). This would create crude ingots, similar to those found on the seabed of the Erme estuary. The ingots produced by the Indian tribes in Babu’s study (Fig.6) were created using simple stone moulds like those in medieval Britain and are much more aesthetically pleasing than the Erme estuary ingots. This suggests that perhaps simpler methods were used for the production of the Erme ingots.



Figure 6. Tin Ingots produced by Indian tribes using simple stone moulds (Babu 2003, ?).


2.4 Chapter Summary


This chapter has demonstrated how prehistoric communities in Britain, with the right resources could have produced tin ingots and also probably bronze. Cornwall’s importance to the tin industry has long been known, out of all the known sources of tin in the world Cornish tin is said to be second only to that of the Malay Peninsula (Thibault 1908, 10). With this in mind it is surprising that there is not more evidence for its extraction in the past.


However the lack of evidence for tin extraction most likely indicates the methods used in prehistory and the rarity of metalworking debris could display the purity of European ores which would produce little or no slag, and perhaps also is a result of the acidic soils such as those on Dartmoor which destroy metal artefacts (Price 1993, 268).


Casting tin ingots could be done by using simple clay moulds in prehistoric times, which again would leave little evidence in the archaeological record.


Although ethnographic evidence is able to assist in the interpretation of prehistoric extractive and smelting techniques, it should be used sparingly as it is most likely that varying regions would each use different techniques and also modern day tribes may be influenced by modern standards.


The interpretation of prehistoric metalwork may also be biased as modern ideas are often very short sighted and do not allow much interpretation to take place. This has been demonstrated by the Indian case study where temperatures of 600ºC were used rather than today’s accepted smelting temperature for tin of 1000ºC.


3.0 The Erme Estuary and Tin Ingots


3.1 Background to the Erme Estuary


The Erme estuary (SX 607 465, Fox 1995, 11) flows down from southern Dartmoor and opens up into Bigbury Bay (Fig.7). The approach to the area resembles that of nearby Plymouth Sound, and has enticed a number of ships over the years. Across the mouth of the river lies a reef of rocks known as East and West Mary’s Rocks (Fig.9) which are just below surface at low tide and have caused numerous ships to flounder.



Figure 7. Location of the Erme Estuary, Devon (ref).


A number of artefacts, including cannons and swivel guns, have been found at the mouth of the Erme by the South West Maritime Archaeological Group (SWMAG). These artefacts are so varied in style and date that they are believed to be from a number of different vessels that have fallen at the site in the past. This site at the Erme became known as the Erme Cannon site (Fig.8) and while waiting for an historic wreck licence to come through, the group decided to explore the end of West Mary’s Reef (Pers. Comms. Neville Oldham) and made a remarkable discovery of tin ingots. At present they could be dated anywhere from the Bronze Age to the medieval period (Cunliffe 2001, 305). To date this has been the largest underwater find of archaeological tin in Europe (Pers. Comms. Neville Oldham).


Figure 8. The positions of both the Erme Cannon Site and the Erme Ingot Site and also the location of East and West Mary’s Rocks (SWMAG 2006).


Figure 9. East Mary’s rocks, displaying the danger a reef can have on an un-suspecting sailor.


3.2 Background to the Recovery of the Tin Ingots


A total of forty-four tin ingots (most complete and some broken fragments) were recovered from the seabed of the Erme Estuary by the South West Maritime Archaeological Group (SWMAG) (Fig.10). The total weight of these is 84.67kg, (although impurities such as stone and the build up of material underwater may have affected this) which would be a substantial amount of tin to a Bronze Age community if this date is proven to be correct.


The site plan below (Fig.11) shows the location of the majority of the ingots found at the Erme Estuary Ingot Site. They each lay on the landward side of West Mary Rocks at a depth of 7-10m (Fox 1995, 11). The relationship between Mary’s Reef and the accumulation of the ingots suggests this to be the cause of the wreckage; however this will be discussed further in Chapter Four.


The ingots were discovered using an underwater metal-detector and after seven had been raised to the surface the sites importance became clear (Fox 1995, 11), leading it to be well recorded and surveyed by SWMAG, and later protected by the Protection of Wrecks Act 1973. The site was surveyed using a main baseline running between the two extremities of the site and two 90º lines triangulated from this to form a grid of 28m x 15m. The group ensured that the area was well covered and each baseline was marked at 2m intervals for accuracy. All visible artefacts were plotted along with natural features and any ‘hits’ on the metal detector were noted. Many of the ingots were accumulated together, shown by the crosses in the site plan (Fig.11). After being adequately recorded in situ, the ingots were all raised for further analysis (Fox 1995, 11-12).


Figure 10. Members of SWMAG raising tin ingots from the Erme (SWMAG 2005- 2006).


Previous studies on the ingots have been limited. One fragment, number 28, was sent to Professor J. Donaldson and Miss M. Trivedi from Brunel University, both of whom are experts on the corrosion of tin. They reported that the ingot appears to have been produced using a slow cooling process due to its high crystallinity and the metal proved to be relatively pure under X-ray microprobe analysis. This is also supported by elemental analysis on ingot numbers 3 and 5 by Dr. Salter of Oxford University which showed the ingots to be approximately 98-99% pure tin (Fox 1995, 14-16). Analysis also showed that the ingots were ß-tin which has a high density of SG-7.28 (Tylecote 1990, 51). Dating of the marine growths present on the ingots was carried out by Camborne School of Mines and provided a date of c.1000BC (Pers. Comms. Neville Oldham), however more work is needed to gain a more accurate date.


Figure 11. Site plan of the Erme Estuary Ingot Site (SWMAG)


3.3 The Ingots


The Erme Estuary collection contains three different styles of ingot. Two are knuckle-shaped (Fig.12) and it has been suggested that there may be a possible third (Fox 1995, 21) (No 36, Appendix C, Plate 4). Three are square or rectangular slabs (Fig.13) which have relatively well defined edges, and the remaining are Plano-convex in shape (Fig.14) with the exception of some small fragments and broken pieces and one Plano-convex shaped ingot with a hole in the middle (See Appendix C, Plate 3). This variety in style may suggest that each was produced by different smelters from (Fox 1995, 21) different communities and brought together to be traded with other regions; this theory will be discussed further in the end Discussion.


All of the ingots vary in size and weight (Appendices B & C) suggesting that there was no pre-chosen weight requested by prospective traders (Fox 1995, 21) and also that they were produced in one-use moulds which were broken at the end of each production for removal of the ingot and this would support the lack of mould finds in the archaeological record for this period.




Figure 12. Photographed by E. Loughman.


The H-shaped or knuckle-shaped ingots (Appendix B, Plate 2) are relatively small in comparison with the other ingots and they are also the most unique in style.  


In his text Bibliotheca Historica, written in the first century AD Diodorus Siculus writes ‘Then they work the tin into pieces the size of knuckle- bones’ (Cunliffe 2002, 76). It is unclear whether or not he means also the shape of knuckle-bones but either way the resemblance between his description and the ingots from the Erme should not be ignored. This description would place the Erme Estuary Ingot Site in the Iron Age; however the Britons may have been producing tin in this style for hundreds of years before this date and until further evidence becomes available this will remain unclear.


Rectangular/Slab Shaped


Figure 13. Photographed by E. Loughman


There are three rectangular and square-type styles in the collection. They resemble the slab types recovered from the Cape Gelidonya wreck and the Ulu Burun Wreck, discussed below. Each has well defined edges and the selection varies considerably in size and weight.




Figure 14. Photographed by E. Loughman


The majority of the ingots raised from the seabed of the Erme Estuary are Plano-convex in shape but vary considerably in size and weight suggesting the use of various moulds/ bowl furnaces to produce them. This shape is said to be one of the oldest types of ingot, often found in Middle Bronze Age contexts (Muhly et al. 1977, 354).


 3.4 Comparisons with Late Bronze Age Ingots


Through macroscopic examination it is possible to compare the ingots from the Erme Ingot Site with other ingots previously discovered.


For the purpose of this paper, ingots from two well known Late Bronze Age wrecks will be examined and a macroscopic comparison with the ingots from the Erme Estuary will be made.


3.4.1 Macroscopic Analysis


The three styles of ingot from the Erme Estuary compare well to the collection of ingots recovered from both the Ulu Burun shipwreck and the Cape Gelidonya wreck, both situated in the Mediterranean (Fig.23). In these collections there are also three styles of ingot present; Plano-convex, slab types and ox-hide types, with the slab types being comparable to the square/ sub-rectangular ingots from the Erme and the ox-hide ingots although very different in size, resembling the shape of the knuckle-shaped ingots from the Erme. This is more easily viewable in the table below.


Table 1: A Macroscopic Analysis of various ingots from the Late Bronze Age wrecks of Ulu Burun and CapeGelidonya, and the Erme Ingot Site



Figure 15. A selection of the Erme Estuary Ingots on display at the RoyalAlbertMuseum. (Photographed by E. Loughman).


After visiting the Royal Albert Museum where most of the tin ingots from the Erme Estuary site are on display (Fig.15), permission was given for six of the ingots to be taken back to Bournemouth University for further studies to be carried out on them. It was decided that X-ray analysis would be the best step to take primarily as it would allow a more precise shape to be seen.


All of the six ingots (Nos, 8, 22, 23, 37, 41, 43) were X-rayed using a Hewlett Packard Flaxitron Series 43804N X-ray system and a combination of variables in time and KVP exposure. These variables were often dependant upon the thickness of the ingot. The negative image was produced on Agfa medical x-ray film and developed in developer fluid (made with 4 parts to 1 part water) and fixer fluid (4:1 with water), for one minute in each fluid (See Appendix D for further details).


After the first few x-rays were taken it became apparent that the shape of the ingots did not alter under x-ray analysis (Fig.16) meaning that they are almost entirely composed of metallic material.


Figure 16. X-ray analysis showed that the ingots are covered by a thin layer of tin oxide (SnO²), the main corrosion product of tin which preserves the original shape.


Further research was then undertaken and it was discovered that the main corrosion product of tin is tin oxide (Cassiterite). This product usually forms a relatively thin layer over the surface of the metal (Tylecote 1990, 51) and also acts to preserve the original shape (Hobbs et al. 2002, 53). This transformation from tin (Sn) to tin oxide (SnO²) can result in large blisters on the surface on the metal due to the volume of the oxides being larger than the volume of the metal from which they form (Tylecote 1990, 51). These blisters (Fig.17) are evident on each of the ingots studied. The Falmouth ingot of tin which was found in salt water had similar corrosion products to the Erme estuary ones, including large blisters and cracks (Tylecote 1990, 51).


Figure 17. Ingot Number 41 displaying large blisters due to corrosion (Photographed by E. Loughman).


3.6 Possible Methods of Extraction and Production


Analysis of the ingots has demonstrated how such research can assist in gaining more detailed information about the formation of such artefacts. It would appear that the varying types of ingots available from the Erme Estuary site may be due to the production of them and also possibly how and where they were extracted from. Pure Cassiterite is often coloured by iron and so it is yellow or reddish-brown in colour or sometimes close to black (Atkinson & Atkinson 1996, 58). The selection of ingots in this collection are mainly black in colour and also have a greasy texture which indicates that they are from water-worn stream-tin (Atkinson & Atkinson 1996, 58). However a small number of the ingots, noticeably the larger ones, are reddish-brown in colour (Fig.18) and have a different texture which may indicate that they are from open-cast workings of tin. Both these methods of tin extraction are known to have been familiar to Bronze Age communities and also to have taken place on Dartmoor.



Figure 18. Ingot Number 6 which is very different in shape and colour to the other ingots suggesting that it may be from a different source. Photographed by E. Loughman.


The difference in colouration however, could also be due to the type of oxides formed during corrosion (Hobbs et al. 2002, 24) and perhaps is due to the deposition of the ingots under water (i.e. how deep they were and how well covered). This needs to be investigated further at a later date.


Examination of the crystallisation of the ingots allows one to gain an insight into the cooling processes involved in casting the ingots. As liquid metal transforms into a solid the crystals form and the size of these will be determined by how quickly the metal is allowed to cool down (Fig.19). Fast cooling creates small crystals and slow cooling creates larger crystals (Palmer & Easterbrook 1999, 7). Areas of the ingot that have been more exposed will cool at a faster rate. The high crystallisation of the tin ingots from the Erme estuary suggests that a slow cooling process was used in their casting (Fox 1995, 14).


Although one can only speculate at the methods of casting used to produce these ingots the evidence available would suggest that they were cast in a warm environment where cooling would be slow. This could indicate that the majority of the ingots are the primary product of the smelt in a small bowl furnace however the rectangular and H-shaped ingots must be a secondary product. Further research should include microscopic analysis of the ingots as the working history e.g. annealing, casting the object, is often visible in the crystal structure (Hobbs et al. 2002, 13) and to date only one of the ingots has been analysed by specialists.


Figure 19. Diagram to show the formation of crystals dependant upon cooling processes (Palmer & Easterbrook 1999, 7).


3.7 Chapter Summary


Studies of the ingots suggest an early date, relative to the style and suggested method of casting used to produce them. Medieval moulds for casting ingots found on Dartmoor show a very precise shape that the ingots would have formed, most are rectangular, 30 x 20 x 12cm(Greeves 1984, 10) and the ingots from the Erme are nowhere near as uniform as this which also suggests an earlier date.


However Plano-convex ingots have a long history and so an early date should not be assumed on this factor alone.


The similarity of the ingots with those from the Late Bronze Age wrecks at Ulu Burun and Cape Gelidonya, especially the Plano-convex and Slab shaped ones should be noted.


Although the ox-hide ingots are thought to be functional, produced for the ease of transportation, either by two individuals or perhaps carriage by horse (Muhly1985, 289) (Fig.20), it is possible that British communities were attempted to copy the renowned ox-hide shapes on a much smaller scale and thus produced the H-shaped ingots.


Figure 20. An illustration demonstrating how ox-hide ingots such as that found in Falmouth harbour may have been transported (Muhly 1985, 289).


As not all of the Erme ingots have been studied, it is possible that they may vary in their elemental make-up. The strong contrast between the majority of the ingots and those that are irregular in shape and reddish in colour may suggest that the larger ingots (Appendix C, Plate 5) were being traded as pure Cassiterite and had not been smelted. This would save the community time and effort and the material could still be highly valuable and may have been smelted elsewhere. Metalworking evidence from south-east Britain (Darvill 2002, 112) and tin ore finds at Hengistbury Head (Cunliffe 2001, 403) shows that ores (which were not found in South-East England) were sometimes traded.


4.0 Bronze Age Trade


4.1 Introduction


Before the discovery of bronze, trade most probably would have been an unorganised and sporadic event. Neighbouring communities would have traded with each other for basic necessities or aesthetically pleasing ceramics and jewellery which one had and the other wanted (Cleland 1927, 234). The introduction of bronze meant that day to day life was made easier; trees could be cut down in half the time it took using a stone axe, cultivated land could be increased, and bronze spears and fishhooks made hunting an easier task. With this in mind, bronze, and the metals used to produce it, must have been much sought after materials which any man would go to lengths to obtain as it would be highly beneficial to him (Cleland 1927, 236).


The art of sailing or rowing a boat was already well established by this time and the sea had already been ventured into by the coastal communities who depended on fishing for much of their daily diet. Therefore if a community needed bronze and lacked the ores or relevant skills to produce it, they would need to go in search for it. This would most probably have been done by boat as it was often the quickest and most efficient way to travel. Once a distant land had been discovered, trade routes could be established and as long as the bronze-lacking community had something of equal value to trade with the other group, exchange could be relatively simple.


It is believed by some (Lyons et al. 1912, 202) that Britain learnt about metallurgy through trade with south-east Europe and Cyprus, however this remains uncertain.


After the discovery of bronze land routes became more widely established and rivers were used as the main source of transport from one region to another, in many ways this was safer than using the sea as coastal boats were smaller and easier to maneuver. Once a community had bronze they could also begin to produce bigger and better boats which may be able to transport larger quantities of goods and hence increase a community’s economy.


4.1.1 The Need for Tin


If one region possessed tin ores but lacked copper ores then it may be possible that each region would have produced ingots of the metal which they contained and shipped them to other regions where they could be alloyed together to produce bronze. Cassiterite (tin ore) is a relatively rare ore (Fig.21) which would increase the value of tin and people may have travelled long distances to find good quality metal.


In the Early Bronze Age Spain was the principle source of tin, however later tin deposits in Cornwall were discovered and became the primary area for good quality tin, causing Spain’s status to decrease and Cornish tin to be used throughout Europe for more than three-thousand years (Cleland 1927, 233).


Figure 21. The location of Europe’s tin ores and major tin trade routes (Cunliffe 2001, 303).











4.1.2 Trade Centres Abroad


Throughout the Bronze Age the eastern Mediterranean became the centre of European culture and this was where most trading took place. Cleland (1927, 235) suggests that in the Early Bronze Age traders exporting tin and copper from Britain would have used sea routes, following the coasts of France, Spain and Portugal, reaching the Mediterranean through the Gibraltar Strait (Fig. 22) as this would have been the familiar way to travel.


Figure 22. Map showing various prehistoric trade routes including over-land and sea-routes (Cleland 1927, 235).


Later over-land routes, using river systems and relatively small river vessels were established and this meant that cargoes could be transported from northern France down to the Mediterranean (Fig.23) meaning that the Cornish tin would only have had to cross one sea, the English Channel, in order to be exchanged in the centre of European culture at this time (Cleland 1927, 236).


Figure 23. Map to show the centre of European Trade and the routes that British vessels may have taken (After Merritt).


Crossing the English Channel should not be considered lightly. As Muckelroy (1981, 279) rightly stated, ‘the sea is by no means so simple and homogenous, and cannot be dismissed so easily.’ The vast area of water between the English and the French coasts is a dangerous environment, open to powerful south-westerly winds and strong tidal streams it is one of the most difficult stretches of water in the world. It should also be noted that in the Bronze Age we may have seen a very different channel than we see today, with changes in climate and seabed topography and also more unstable sediment patterns and possibly stronger currents as a result of the closeness in this time to the land-bridge between Britain and the Continent being broken (Muckelroy 1981, 280).


Shipwrecks including the Uluburun and Cape Gelidonya, and perhaps now the Erme Ingot Site, demonstrate how ingots of copper and tin were being exploited during the Later Bronze Age. Both the shipwrecks found in the Mediterranean were large vessels carrying immense cargoes in the European trading centre, this is very different to the Erme site which at present appears to have only been carrying a small cargo of tin ingots. This may demonstrate the likelihood that the vessel from the Erme was a small coastal vessel exporting a small quantity of locally manufactured ingots to a nearby trading site which potentially could then have taken them, had they reached their intended destination, to the Mediterranean.


4.2 Possible Trading Sites connected to the Erme


The use of the term ‘Ictis’ by such writers as Diodorus has caused much discussion amongst scholars over the past years. The word was used in reference to the trade of Cornish tin in prehistoric times and although clues are given for the location of the island, it is still yet to be established. The following is part of the quote made by Diodorus (5.1-4) in the first century AD; ‘and convey it to an island which lies off Britain, called Ictis; for at ebb-tide the space between this island and the mainland becomes dry and they can take the tin in large quantities over to the island on their wagons.’ (In Cunliffe 2002, 76). The islands importance seems to suggest that it was already a well established centre for trade by this time which may indicate that it had been in use since for many years before this reference was made.


The description of Ictis made by Diodorus has led to a number of areas to be put forward as potential origins for the renowned Ictis including; Mount Batten, St. Michaels Mount, Burgh Island, Hengistbury Head, and the Isle of Wight, along with others. Mount Batten is often thought of as the most promising site to be connected to Ictis due to its location and also the prehistoric finds that it contains. The close relationship between this site and the Erme estuary (Fig.24) is also worth noting.



Figure 24. Map to show the close relationship between Mount Batten (A) and the Erme Ingot Site (B). (After Fleming 1988, 4).


4.2.1 Mount Batten, Plymouth


Mount Batten lies in the heart of Plymouth Sound (Fig.25), a drowned river estuary some 4km wide which is fed by four of Southern Dartmoor’s great rivers (Cunliffe 1988, 1). Mount Batten is a relatively small outcrop of land, oval in shape, and joined to the mainland by a narrow neck (Fig.26) which could easily have become submerged under high tides before the construction of Plymouth Breakwater in the 1800’s (Cunliffe 1988, 1), a characteristic much like that described by Diodorus when speaking of Ictis.


Figure 25. A map showing the location of MountBatten, South West England.


Archaeological discoveries between 1830- 1982 have included much prehistoric metalwork along with other prehistoric finds such as pottery. Metal tool finds assigned to the Late Bronze Age are thought to date from the Ewart Park – Llynfawr periods (See Appendix A for dates) and interestingly there is no evidence for the earlier Wilburton Period. Tools on site have similar contemporaries across the continent in France which may demonstrate cross channel exchange (Northover 1988, 75-7), and much of this material appears to have been used for scrap which suggests it was being imported or exported for the metal content (Northover 1988, 81).


There is also substantial evidence for metalworking on site. Waste drops and four cakes of metal suggest casting activity. Although these artefacts are not datable an examination of the composition shows that they fit into the Ewart Park and Llynfawr Periods best (Northover 1988, 81, 83). Apart from one fragment of copper ingot all the waste metal material is bronze or leaded bronze and the tin contents were shown to be varied (Northover 1988, 85).



Figure 26. Photograph of MountBatten facing North c.1965. The narrow neckline can be seen and in prehistoric times this may have been covered at high tide (Cunliffe1988, 9).


The lack of personal items on site may suggest that it was a site primarily used for work rather than a settlement. It should also be noted that the only weapon found was that of an arrowhead (Northover 1988, 81), which suggests that the trade was friendly and civilised; two characteristics also mentioned in Diodorus’ text.


Although archaeological evidence is sometimes biased due to the nature of survival of certain finds, it can be assumed that the nature of this site was occupational and that trade connections were mainly with the continent. Most of the materials could have come from either side of the channel which shows the close relationship between the two regions. The evidence for metalworking on site suggests that perhaps as scrap metals were imported they were being melted down and re-cast into new, desirable artefacts.


4.3 Bronze Age Vessels


The discovery of Bronze Age vessels has not surprisingly been somewhat limited. Wooden vessels of any age are often difficult to find in the archaeological record due to the composition of wood which deteriorates relatively quickly in aerobic conditions and also is quickly washed away from the place of wreckage unless heavier items fall on top of it during deposition.


This chapter hopes to examine the various types of Bronze Age vessels in order to make assumptions on the type of vessel that may have carried the tin ingots to the Erme Estuary site.


There have been a few major discoveries of Bronze Age vessels which have helped to highlight the use of ships and boats in this period for travel and trade (See Appendix A).


The relatively small cargo of the site may suggest that the tin ingots recovered from the seabed of the Erme estuary were being transported around the coast, either from a trading site such as Mount Batten or perhaps from the Erme itself to a trading site.


4.3.1 Logboats


During this period logboats were often used and they were in some ways the easiest vessel to produce as no stitches were needed and there was less chance of leakage. However evidence for such craft indicates that they were fairly large in size (Brigg logboat = 14m, Hasholme logboat = 13m) which suggests that they would not have been suitable for river transportation (Roberts 2004, 40). A vessel such as this could have been carrying the ingots but this would rule out the theory of the ingots coming down the River Erme unless a smaller type was used.


4.3.2 Hide Boats


Hide boats were built frame first and then covered with a hide, which was fastened to the framing. This simple design would not have required the use of many tools and could potentially have been the earliest type of craft used by man (McGrail 2004, 63). It is probable that these vessels were used in N.W Europe for fishing, ferrying and carrying cargo in the Upper Palaeolithic/Mesolithic periods and larger varieties (Fig.27) most probably would have been available in the Bronze Age (McGrail 1998, 186). Its seaworthiness and stability meant that it could have been used to make voyages across vast quantities of sea, and its small size made it easily accessible for river transportation. Potentially the hide boat could have been ‘the workhorse seaboat of prehistoric north-west Europe’ (McGrail 2004, 65). Its use has been documented since 55 BC when the Romans invaded Britain (Blandford 1974, 22); however a distinct lack of archaeological evidence for such craft means that one can only speculate about its use in prehistory. It should be noted that such a vulnerable craft would rarely survive into the archaeological record, unless in peat deposits and a lack of archaeological remains does not necessarily mean they were non-existent.


Figure 27. Could a vessel such as this Hide boat have transported the tin ingots around the coast of Devon? (Blandford 1974, 27


4.4 Chapter Summary


No-one knows the true identity of Ictis but archaeological evidence on a number of sites such as the Early Bronze Age continental metalware at Hengistbury Head (Williamson 1998, 6) suggests that there would have been more than one trading centre along the south coast during the Bronze Age.


The archaeological record for Bronze Age vessels is comprised mostly of river and coastal vessels, (demonstrated by the distribution of sewn-plank boats around Britain, Appendix F), which rather than indicating that the people of the Bronze Age did not venture into the open sea demonstrates that sea-going vessels have simply not survived into the record or are yet to be discovered. This supports the theory for the Erme Ingot Site being a wreck site, as in such a strong tidal area it would not take long for a small vessel to be ravaged by the sea.





5.0 Wreck site versus Ritual Deposit


5.1 Introduction


Due to the lack of evidence for ship or boat structure at the Erme Estuary Ingot Site some have suggested that it may be a ritual deposit, thrown to the gods as a sacrificial offering in the hope that it would bring the local community luck, while others argue that the lack of actual wreck remains does not necessarily mean that it is not a shipwreck. Other sites such as Langdon Bay and Moor Sands have brought up similar problems, but are interpreted as shipwrecks due to other evidence available. This chapter will focus on shipwrecks such as the Uluburun and Cape Gelidonya from the Mediterranean, as well as the more local sites of Langdon Bay and Moor Sands in order to compare each with the Erme Ingot Site. Ritual deposits will also briefly be examined.


5.2 Bronze Age Wreck Sites


The following four sites have all been dated as Middle or Late Bronze Age wreck sites. Both the Ulu Burun and the Cape Gelidonya sites are important to this paper as they are wrecks which held large numbers of copper and tin ingots in their cargoes. The Langdon Bay and Moor Sands Sites did not contain any ingots but were carrying large quantities of bronze artefacts, presumed to be scrap-metal, and are important in relation to the Erme as they have been interpreted as wreck sites even though no actual vessel structure has survived. (See Figure 23 for locations).


5.2.1 Cape Gelidonya


The Cape Gelidonya shipwreck was discovered in 1958/9 lying between two islands close to the mainland of south-west Turkey (Bass 1961, 267). Copper ingots formed the largest part of the ships cargo (Fig.28).


Figure 28. Plan of the CapeGelidonya wreck site (Bass 1961, plate 85)


Interestingly, most of the objects were broken and were transported alongside fragments of ingots suggesting that it was the metal, not the artefact that was in demand (Bass 1961, 274). Other tools were kept in good condition and whetstones were onboard to sharpen them. Forty-eight weights, consisting of three sets were also carried onboard reinforcing the fact that this was a merchant’s vessel (Bass 1961, 274).


Some of the structure of the ship survived but it was poorly preserved in just a few centimetres of sand. The planks were fragmented and dowels were found in some holes. The ship has been estimated as being 8-9m in length due to the distribution of the heavy cargo (Bass 1961, 271). This would have been a moderately small vessel, perhaps solely used as a coastal vessel in the relatively calm waters of the Mediterranean. Although the width is unknown, the minimum dimensions calculated provide a small ship that is perfectly capable of carrying the cargo of approximately 1000kg along with the additional weight of the ballast stones (116kg), (Bass 1961, 271).


The ship has been dated to the twelfth-century B.C. and it has been suggested that it was carrying copper from the mines at Cyprus due to the marks on some of the ingots being identical to marks on pottery from Cyprus and the Cypriot colony in Syria. The ship was therefore most likely to have been travelling from east to west where the ingots could be exchanged for other desirable items.


5.2.2 The Ulu Burun


The Ulu Burun is a Late Bronze Age shipwreck, discovered on the southern coast of Anatolia in 1982 (Hauptmann et al. 2002, 1). In its cargo it held an astonishing ten tons of copper ingots and one ton of tin ingots; it should be noted that, perhaps by chance, this is the same ratio of copper and tin needed to make classic bronze (Hauptmann et al. 2002, 2).


As discussed earlier (Chapter Three) the Mediterranean was the centre of European culture at this time and the range of onboard materials including ivory, glass and frankincense suggest that this was a substantial ship and an important merchant vessel of the time carrying goods around the Mediterranean, possibly from east-west along the coast as supported by other wrecks (Bass 1986, 270).


Some of the hull structure remains on the seabed, mainly due to the heavy load of ballast and other heavy goods resting on top (Bass 1986, 275) and the positioning of the ship further supports the east-west movements of the merchant vessel. Mortise-and-tenon joints appear to have been used for much of the construction including the attachment of the garboard strake to the keel and a secondary strake (Bass 1986, 275). This technique was common in antiquity (Gardiner 2004, 136).


The wreckage covers an area of approximately 10m x 18m and the site plan below (Fig.29) shows how the ingots were arranged in rows, stacked on top of each other with ballast stones and small artefacts lying between them and the stone anchors. The heavy items appear to lay athwart-ship in groups (Bass 1986, 274), possibly to help stabilise the ship




Figure 29. The site plan of the Ulu Burun wreck (Bass 1986, 273).


The ceramic evidence onboard suggests a date of the fourteenth century B.C. and the ingots of ox-hide shape further support this, as they were commonly in use around the fourteenth – late thirteenth centuries B.C. The Plano-convex shaped ingots are said to be earlier (sixteenth- fifteenth centuries B.C.) however these were likely to have still been in use around this date.  


5.2.3 Langdon Bay


In 1974 a group of divers from the Dover Sub-Aqua Club (Parham et al. 2006, 43) discovered a hoard of metal artefacts at Langdon Bay (Fig.30), 500m seaward from Langdon cliffs at a depth of 7-13m (Muckelroy 1981, 275). The site has been dated to the end of the Middle Bronze Age (1300 B.C. Parham et al. 2006, 43) and the collection, comprising of some 360 artefacts, makes it the biggest mass of metalwork in north-west Europe for the period (Fig.31) (Parham et al. 2006, 43).



Figure 30. Location of LangdonBay in South East England (


Most of the finds are concentrated within a 50m radius of the centre of the site and although no actual shipwreck remains have been discovered it has been interpreted as a wreckage site rather than a ritual deposit due to a number of affecting factors; the quantity of finds, the position of finds including their concentration, and the French character of the metal artefacts. The majority had been cut, thought to be for convenience reasons when transporting them (Muckelroy 1981, 275), and comprised of bronze tools, weapons, fittings and ornaments (Parham et al. 2006, 43). The exceedingly large number of artefacts found at Langdon Bay is unparalleled and outnumbers the total known artefacts of this type in Switzerland alone, demonstrating how one maritime site can cause an imbalance on a distribution pattern of artefacts shown on land (Muckelroy 1980, 101).



Figure 31. The hoard of metal artefacts recovered from LangdonBay on display at the DoverBoatMuseum (Photograph D. Parham).


5.2.4 Moor Sands


This site is comparable to the Langdon Bay hoard in its contents but is far smaller. It was discovered in 1977 just off Moor Sands (Fig.32) to the west of Gammon Head (Parham et al. 2006, 43), and some 3km east from the mouth of the Salcombe estuary (Muckelroy 1981, 276). A total of eight artefacts have been recovered from the seabed; six swords and two palstaves (Parham et al. 2006, 44). The first two finds were discovered completely exposed on top of a gravel seabed approximately 3-5m apart and the lack of encrustation and obvious exposure suggests that the two artefacts had been deposited together from a stable environment not long before the find and washed inshore along with the tide (Baker & Branigan 1978, 149). Following this first discovery further surveillance of the area took place in the two consecutive years, led by Keith Muckelroy and Philip Baker. These investigations helped to further interpret the site. Four finds were recovered in the 1978 season and just one bronze was produced in 1979 (Muckelroy & Baker 1980, 155-156).


The importance of the site lead it to become protected under the Protection of Wrecks Act 1973 in March 1978 (Muckelroy & Baker 1979, 189).


Although this is a much smaller content than that of Langdon Bay and could be regarded as a small collection of items deposited for ritualistic events, it is often interpreted as a shipwreck site due to certain factors. The nature of the finds suggests that they had not long been released from a stable environment and if this environment was that of a cliff deposit then one would expect to have see evidence for recent (in the 1970’s) cliff erosion, for which there was little (Muckelroy & Baker 1979, 204). Also it is doubtful that the delicate composition of the sword with its fragile hook-tang would have survived a violent fall from a cliff.




Figure 32. Location of Moor Sands in the South West of England (After Fox 1995, 12).


The French character belonging to all the found bronzes also provides some argument against either ritual deposition or cliff deposition. Why would a collection of restricted French material be deposited with no other materials? In this sense it is much more likely that a ship carrying continental artefacts was wrecked in this region.


The sites relationship with the Salcombe B site discovered by the South West Maritime Archaeological Group (SWMAG) also provides evidence for a shipwreck. This site lies just 500m west of the Moor Sands site and twenty-eight items of probable Bronze Age date, contemporary with the those from Moor Sands, have been recovered (Parham et al. 2006, 45). The close relationship between the two and the exposed conditions of the area suggests that they were in fact originally from the same wreckage (Pers. Coms. Neville Oldham). When examining the distribution of artefacts on Moor Sands it becomes clear that the lower mass items are found closer to the shore and the heavier items are further out in deeper water (Muckelroy & Baker 1979, 205). This suggests that the artefacts have been washed inshore and could potentially have been originally from one site.


However the wide dispersion of the artefacts is little understood at present and so one should aim to keep an open mind on this fact until new information becomes available (Parham et al. 2006, 45).


5.3 Ritual Deposits


Ritual deposits largely consist of materials which cannot fit into the terms of utilitarian deposits and also vary between regions, making them difficult to distinguish (Needham 2001, 278). For the purpose of this paper, ritual deposits will form the focus, as the location, condition and arrangement (elements used to define character of depositions (Bradley 1985, 693)) of the tin ingots from the Erme Estuary are more likely to resemble ritual deposition than utilitarian, if any.


Ritual deposits, or votive offerings, are usually sacrificial forms of deposition related to ceremonial rites of passage, made at a time when a community is in need of something or wishes to have luck (Bradley 1988, 252).


It is quite common for items of ritual deposition to be discovered in watery locations, such as lakes, rivers and marshes (Bradley 1985, 693). River finds tend to be more important than bog finds and they become more common towards the Middle/ Late Bronze Ages (Needham 1988, 230). The ingots of tin found in the mouth of the Erme Estuary however do not seem to parallel the finds of votive offerings in other watery deposits. These are usually valuable artefacts which have been skilfully produced and are extremely prestigious. Although the ingots most certainly would have been a valuable commodity due to their metal content, they are not prestige goods and it is unlikely that they would have played a role in a sacrificial deposition to the gods. It should also be noted that while hoards often contain copper and bronze, tin is rarely found amongst them suggesting that its supply was highly organised and valued (Rohl & Needham 1998, 15).


5.4 Chapter Summary


When the Erme Ingot Site was surveyed no trenches were made and metal detectors formed the basis of the surveillance. This means that artefacts such as stone anchors, pottery, and vessel remains may actually exist on the site but are yet to be discovered. The lack of iron finds however which would contribute to a wreck of Iron Age or later date suggests a Bronze Age date for this site.


The accumulation of ingots on both the Uluburun and Cape Gelidonya wrecks is comparable to the accumulation of ingots at the Erme estuary (Fig.11) which may support the idea of it being wreckage rather than an area of ritual deposition as one would expect a higher degree of distribution at a deliberate depositional site. Although ritual deposition should not be ruled out entirely the nature of this deposit does not seem to resemble such a site.


The problem of the nature of deposition is one that is common among Bronze Age remains, for example the Shardlow logboat, which can be interpreted as either a wreck carrying too much cargo or a vessel which was deliberately sunk (Pryor 2004. 33).


Such remains are often open to interpretation.


Both the Cape Gelidonya and Langdon bay sites contained scrap metal which demonstrates how metal was re-used during the Middle/ Late Bronze Age and traded to areas that needed it.


The discovery of a Sicilian object from the Salcombe B site and the French character of both the metal from the Langdon Bay site and Moor Sands support the evidence for trade with the continent and the Mediterranean from the south coast of England in the Bronze Age.


The change in sediments on the Moor Sands site was exhibited over the two seasons, showing how these can have an effect on the archaeology of maritime sites and their subsequent uncovering of artefacts after thousands of years on the seabed.


The distribution of artefacts in relation to their weight demonstrates how the environment of the sea can re-arrange stratified contexts meaning that in fact there are often no reliable stratigraphic sequences under the sea.


The density of the tin ingots from the Erme Estuary may suggest that while they have stayed close to the origin of deposition, other artefacts that may have been deposited at the same time could have been washed inshore towards the Erme Cannon Site where there is a varied selection of material. Further research in this area is therefore needed


6.0 Bronze Age Settlement Sites and the Erme Ingot Site


6.1 Introduction


The South West of England, notably Devon and Cornwall, is well known for the quality of tin that it comprises; and the Erme estuary, situated in south-west Devon is a prime location for ancient tin. There are many known Bronze Age sites surrounding it and although the wreckage location of the Erme Estuary Ingot Site should not necessarily be considered the place of origin, the nature of the ingots suggests that they had been produced locally and the quantity (c.84.67kg. Fox 1995, 17) suggests that they were on their way to a trading site where they could be exchanged for other goods. With this in mind, and taking into account the high tin value of Dartmoor it could be suggested that the ingots were coming from this area.


6.2 Dartmoor


Dartmoor (Fig.33) is an area that has been largely exploited for its resources in the years gone by. It is said to be the ‘most completely preserved landscape of the second millennium BC in Europe’ (Collis et al 1984, 1). It is a complex settlement site consisting of 274 or more prehistoric settlements of various sizes (Price 1993, 261).




Figure 33. The location of Dartmoor in the South West of England (Collis et al. 1984, 2).


Ritualistic monuments including cairns and round barrows belonging to the Bronze Age (Greeves 1984, 2) have caused much discussion in previous years and the extensive reave systems (Fig.34) that exist have broken the area into territories (Fleming 1978, 103) each with their own set of settlements and communities.


Figure 34. The extensive reave systems of Dartmoor (Fleming 1988, 59).


These reave systems dating to the Early Bronze Age (Fleming 1988, 95) appear to post-date the first settlements on the moor (Darvill 2002, 109) suggesting that something was happening at this time which required improved organisation.


The majority of settlement sites are characterised by enclosed hut groups (Pettit 1974, 37) and each larger territory contains a number of settlements which accommodate numerous huts. The Avon is the largest territory in southern Dartmoor consisting of twenty-nine settlements, followed by; Plym twenty-four, Erme twenty, and the Yealm eight (Pettit 1974, 37-50). These four are the largest in southern Dartmoor, accounting for three-quarters of all known settlements (Pettit 1974, 37).


There are also a few settlement sites which appear to represent small villages in their complexity (Fig.35) and it has been suggested that these were local centres (Bewley 2003, 84).



Figure 35. Aerial Photograph showing a complex Bronze Age settlement pattern on Dartmoor (Bewley 2003, 85).


Most of the settlement sites appear to have been used for pastoral activities; however it has been suggested (Price 1993, 261) that it would be highly unlikely that pastoral work was the only occupation carried out by the Bronze Age communities.


6.2.1 Evidence for Tin Extraction


Around 2500 BC there is evidence for the land being under considerable new pressures (Fleming 1988, 95). This is supported by pollen analysis, the construction of reaves, and the density of walled settlements, all dating to this period, representing high exploitation of the area. This is around the same time that bronze was becoming more widespread and could be related to Dartmoor’s accommodation for metal. The evidence suggests that the previous seasonal occupation of the land was by now becoming more permanent which may support the theory of tin extraction being used as a means of making a living through the winter.


Tin, copper, silver, lead and iron have all been mined on Dartmoor in previous years and tin is by far the most important of these (Greeves 1984, 10). However this material is also one of the most difficult ores to recognise (Reid 1918, 10). Much of the work was carried out during the medieval periods and waste heaps connected to streamworks have built up in many areas, particularly in the Plym, Erme and the Yealm (Greeves 1984, 10). Although these are evidence of later workings it is possible that these areas were also the main sources of tin for earlier workings.


The major evidence for Bronze Age tin extraction on Dartmoor is the relationship between the settlement sites and streamworks (Fig.36) (Price 1993, 261), a factor that cannot be ignored given that tin was such a scarce and valuable material and hence would have provided great economic benefits to a small community.




Figure 36. The relationship between streamworks and larger settlements on Dartmoor (Gerrard 2000, 18).


Evidence has shown that each territory has three components; access to higher moorland, access to lower moorland usually associated with a river valley, and a parallel reave system (Fleming 1978, 104). It is common for settlements to be built on or around a river system as every village needs a source of water; however the relationship between settlements and known streamworks is a tantalising one.


Most huts form clusters in areas associated with farming activities, however there are some huts which are found to be isolated, sometimes more than 150m from other huts. Very little is known about these buildings, and sometimes they are isolated due to topographical features that cause obstructions (Price 1993, 264), however their existence should not be ignored. It has often been discussed that the people working the metal, the smiths, would in some societies be kept separate from the rest of the community because of the awesome powers that they may possess (Parker Pearson 1993, 84). This often depends upon regional variety, however due to the abundance of farming on Dartmoor it is likely that in this case, smiths were part of normal society and carried out both farming and metalworking activities dependant upon the seasons (Rowlands 1971, 213).


Gerrard (2000, 19) suggests that smelting sites were not always confined to settlement sites and the location of a building next to streamworks would have its advantages in that transportation of materials would be minimised. On Dartmoor two of five huts over 1000m from enclosures form a close relationship to streamworks (Price 1993, 264). This could be due to the disturbances of later streamworkings causing other buildings to have been demolished, however no interpretations should be ignored until new evidence is able to disprove it.


Price (1993, 266) suggests that the contrast between various settlement types may be related to function; for example settlements relating to field systems may be related to farming activities. It is interesting that narrow, linear settlements with little room for farming are situated near river systems; perhaps this indicates that such settlements were carrying out tin extractive activities?


The largest settlements occur in areas which do not provide the best conditions for farming and instead are situated close to known areas of tin exploitation. These areas consist of deposits related to erosion during the last glacial period causing tin ground to rest on top of solid rock which could have been removed by prehistoric tinners (Price 1993, 272). Settlement patterns (Fig.36) show that the larger settlements are more closely related to streamworks than the smaller settlements, with over half of the larger ones being within 250m of them.


Environmental evidence for the prehistoric mining of tin on Dartmoor can now be used to back this up. Tin streaming releases large quantities of sediment waste which are washed down-stream and redeposited into the river valleys below. High quantities of tin are often found in these deposits and can thus be measured. Two layers of peat, sealed by alluvial sands and silts were discovered in the ancient channel in the floodplain of the River Erme and the layers of peat were radio-carbon dated. The upper layer gave a not surprising date belonging to the medieval period (AD 1288-1389) and the layer of sandy silts above the lower layer was then dated to the Late Roman/Early post-Roman period. More surprising was the dating of the lower layer of peat. This contained sandy silts which had a small but relevant tin content dating to c. 991-850 BC (Devon Archaeological Society 2005, 5) and may demonstrate that Bronze Age communities living on Dartmoor were utilising the tin deposits for their benefit.


6.2.2 Evidence for metalworking


Some metalworking debris has been found on parts of Dartmoor. A piece from eastern-Dartmoor dated to c.1400-900 BC and was associated with the contemporary settlement (Price 1993, 275). Excavations at Dean Moor in the Avon Valley also suggest metalwork was taking place. A Cassiterite pebble and minute piece of tin slag were recovered from two separate huts, and although most metal artefacts had been destroyed by the acidic soils, the presence of whetstones in every hut strongly indicates that metal tools were used (Fox 1957, 30). It was noted that stream tin was available nearby and the slag could not have been produced accidentally (Fox 1957, 31). This, along with the presence of iron ore (mined from an outcrop) built into a wall, suggests that the inhabitants had knowledge of metals but not enough to know how to smelt iron, providing a Bronze Age date for the settlement (Fox 1957, 33).


Although some metalworking debris has been discovered on Dartmoor there is still a distinct lack of it. However this could be due to a number of factors.


  • Dartmoor’s acidic soils can      be very destructive to metalwork and other artefacts (Price 1993, 268).
  • Lack of excavations on      Dartmoor (Gerrard 2000, 19).
  • The low effect      streamworkings have on the environment.


Additional pollen analysis may be able to support the metalworking evidence on Dartmoor during the Bronze Age. A decrease in alder pollen around the time that the reaves were in use (Fleming 1988, 105), and the suggestion that this tree was particularly suitable for the use of smelting (Price 1993, 276) may demonstrate how this tree, often found close to rivers and other wet areas, may have been exploited for tin smelting.


6.3 Chapter Summary


The well organised construction of the reave systems around 2500 BC demonstrate the developing nature of the Bronze Age communities. Such organisation represents a great investment of energy and time and also perhaps political power in order to instigate it (Darvill 2002, 109) and thus suggests that something was to be gained from it.


is situated in a region with good trade links during this period, most noticeably local Plymouth Sound to the west but also around the south coast. The English metal trade with the continent has been shown to be related to Devon in the past through wrecks such as Moor Sands (Muckelroy 1981) and metal artefacts at Mount Batten (Northover 1988). This developed trade may have been controlled by outside forces that may also have held control over areas with high metal potential such as Dartmoor.


Dartmoor was in a central position to local Bronze Age communities (Bewley 2003, 84) and it is possible that its role in the Bronze Age may have been underestimated.


The smaller settlements may have played a role in tin production, and other activities such as farming were probably carried out alongside rather than existing on their own. The more land a community could work meant more income and the nature of mining labour probably meant that a large number of people would have been employed in this area (Price 1993, 278).


Large scale excavation is needed on Dartmoor in order to establish any correlations between settlement sites and the tin industry (Gerrard 2000, 19).


The lack of metalworking finds is not restricted to Dartmoor but is common all over the south-west. Small clues, such as fragments of slag from Cornwall dating to the Bronze Age (Tylecote et al. 1989, 435) and small pieces of metalworking debris, noticeably crucible fragments found in Late Bronze Age contexts at Dainton, Devon (Needham 1980, 181), are provided but much is left up to the imagination as to how such materials were worked.


The relationship between the Erme Estuary ingot site and Dartmoor cannot be ignored. Finding a hoard of crude tin ingots so close to a rare source of tin seems to indicate that these ingots were produced from a local site such as Dartmoor.


7.0 Discussion and Conclusions


7.1 Discussion


From the collected information and study of the tin ingots from the Erme Ingot Site it has been possible to gain a much greater insight into the potential date of the ingots and also the role they could have played in Bronze Age activities.


Through the study of ancient tin extraction techniques and by examining the available evidence for metalworking in Britain it has been possible to establish that it is highly likely that Bronze Age communities in Britain would have been producing and working metal artefacts. The lack of evidence suggests poor preservation rather than non-existence, and the various artefacts that have been recovered and dated to the Bronze Age (for example Fox 1957, 30) provide proof for this.


This information contrasts with previous papers (Barber 2003, 79) which have suggested that British communities may not have been able to produce metals themselves and instead would have been importing metals from other areas such as Ireland.


The study of Bronze Age methods of smelting and casting has led to a wider understanding of how these ingots may have been produced by Bronze Age Communities in Britain. Typological similarities between the Ulu Burun and Cape Gelidonya ingots and the Erme estuary ingots strongly indicate the possibility that these ingots may be of the same period. The varied selection of ingots also suggests this, as in later periods when stone moulds were used, more uniform ingots appear, such as those manufactured in India (Fig. 6).


Wreck sites such as Langdon Bay and Moor Sands have displayed a relationship with continental sources in the Middle Bronze Age, shown also by continental metalware at Mount Batten in the Later Bronze Age. The number of trading centres around southern Britain (Hengistbury Head, Mount Batten etc) also demonstrate how this was an important region of trade during prehistory.


The distinct lack of vessels for this period suggests that they are unsustainable in archaeological contexts, and the lack of remains at the Erme estuary may be a direct result of the environment in which it was deposited or may be due to the fact that the area has not yet been recorded through trenches- wooden structure would not be found through the use of metal detectors alone!


The lack of remains at both the Langdon Bay and Moor Sands sites is also comparable to the Erme estuary ingot site, and proves that it is a common feature of many wreck sites, especially older ones.


A number of other artefacts such as ceramics or stone anchors may to be present on the Erme Ingot Site, however only future work on the site will be able to prove this.


The lack of additional metallic materials may suggest that this was a local cargo being exported to another site. If the ingots belonged to a vessel that had just left a trading site such as Mount Batten one would expect there to have been a greater variety of artefacts, unless such items were perishables such as clothing and skins.


The accumulation of the ingots on site (Fig.11) is similar to that at both Ulu Burun and Cape Gelidonya, demonstrating how heavy artefacts from a wreck are deposited upon wrecking and further suggesting that this is a wreck site rather than one of ritual deposition.


The relationship between prehistoric settlement sites on Dartmoor and streamworkings and the few remains of metalworking debris in this region indicate that it is highly probable that during the Bronze Age the communities living in Dartmoor would have been utilising tin to the best of their knowledge.


Organised networks of trade for this period have been demonstrated by the Ulu Burun wreck where it is believed that royal assignment were sent for the ingots (Bass 1987, 709). This organisation may also be reflected in the construction of the Dartmoor reaves and the complex settlement sites which may have been important centres (Bewley 2003, 84).


After analysing the information on Moor Sands and noting the relationship between that site and the Salcombe B site the author feels that the same interpretation could potentially be given to the Erme Estuary sites. The Erme contains two archaeological sites, the Erme Cannon Site and the Erme Ingot Site (Fig.8). The author feels that the high density of the tin ingots found at the Ingot Site and the fact that they were the sole artefact recovered (although trenches may prove otherwise), may suggest that lighter items have been washed inshore with the tide and deposited at the Cannon Site where there is a large variety of finds which could be from a number of unrelated wrecks (George 1991, 4). The location of the Cannon Site could explain why such a variety of objects exists there as it is in a prime location for items to be washed into and collected.


7.2 Recommendations for future research


It is felt that three key factors will need to be researched further in order to fully understand the Bronze Age tin trade and its possible relationship to the Erme Ingot Site.


7.2.1 Ingot Analysis


Future analytical work could help to establish an origin for the tin ingots and microscopic work may assist in learning more about their working history (Hobbs et al. 2002, 13). Experimental work has also proven to be useful when examining past production techniques (See Appendix E, Tylecote 1987, 196) and could be looked into for the future.


To date, only two of the ingots have been analysed elementally and although this has been useful in gaining an insight into their composition, the author feels that more of the ingots should be analysed to see if they are all the same, perhaps there will be indicators through this analysis which may help to show that they were collected or produced in different ways and perhaps by several small communities.


7.2.2 Excavations at Dartmoor


Although research of Dartmoor has shown clear relationships between settlements and streamworkings (Price 1993) on paper, and small-scale excavations such as that at Dean Moor (Fox 1957) have provided small tantalising clues to metalworking, it is evident that the only way this relationship can be proven is by a large-scale excavation of the settlements and environmental analysis of the tin streams (Gerrard 2000, 19). However this would require a lot of funding and probably will not be undertaken anytime in the near future.


7.2.3 Excavations at the Erme Estuary


To date, the Erme Ingot Site has been surveyed using only metal detectors and so it is felt that other methods of excavation are needed. Plotting trenches on the site will allow non-metallic finds, if there are any, to be found.


The idea that the two sites at the Erme estuary may be related (mentioned above) could be resolved by local stratigraphic seabed research which may help to understand the various sequences of regional seabed activity further.


7.3 Conclusion


As metallic tin is not datable (Tylecote et al. 1989, 435) there is no easy way to discover the true age of these ingots, however through researching related topics and identifying how the ingots may have fitted in to Bronze Age activities it has been possible to conclude that there is a strong possibility that the ingots from the Erme Ingot Site are British Bronze Age in date.


With this in mind the author proposes a theory for how the ingots came to rest at the bottom of the Erme Estuary. It is possible that the ingots were produced locally on Dartmoor in a variety of settlements (explaining the various types e.g H-Shaped, Slab) and were then transported down the river Erme which was much larger in those times (Pers. Comms. Neville Oldham), with the intention of being taken to Mount Batten or a similar trading Centre where they could be exchanged for desired commodities. On the way out of the Erme conditions may have worsened, causing the boat to hit West Mary’s Rocks and capsize and the heavy artefacts (tin ingots) to fall out, sinking quickly and accumulating closely together. Lighter materials may have been washed away or quickly recovered by the crew.


One can only speculate at how these tin ingots came to rest at the bottom of the Erme Estuary site and as with many things in archaeology the sites deposition is open to interpretation. However this paper has highlighted how maritime sites can shed new light on past commerce and activities on land. Such wreck sites as the Erme act as ‘exchange frozen in time’ (Muckelroy 1980, 108) and from them we are able to learn a great deal. This paper has demonstrated how the integration of maritime archaeology into the wider discipline of terrestrial archaeology can provide invaluable information to all kinds of archaeologists. The study of one group of tin ingots from a maritime context has lead to a wider knowledge of prehistoric seafaring and trade, as well as events such as tin extraction and metalwork on land.




Alexander, W., and Street, A., 1989 (9th ed) Metals in the Service of Man. Suffolk: Richard Clay Ltd.


Atkinson, R., and Atkinson, F., 1996 Observers Book of Rocks and Minerals. Revised Edition. Great   Britain: Frederick Warne.


Babu, T.M., 2003 ‘Advent of the Bronze Age in the Indian Subcontinent.’ In P. Craddock and J. Lang (eds). Mining and Metal Production Through the Ages. Pp.174-180. London: The British Museum Press.


Baker, P., and Branigan K., 1978 ‘Two Bronze Age Swords from Salcombe, Devon.’ In The International Journal of Nautical Archaeology. Vol.7, No.2, pp.149-151. The Nautical Archaeology Trust.


Barber, M. 2003 Bronze and the Bronze Age: Metalwork and Society in Britain c.2500- 800 BC. United   Kingdom: Tempus Publishing.


Bass, G.F., 1961 ‘The Cape Gelidonya Wreck: Preliminary Report.’ In American Journal of Archaeology. Vol.65, No.3, pp.267-276. Archaeological Institute of America. Accessed through JSTOR. 23.02.07.


Bass, G.F., 1986 ‘The Bronze Age Shipwreck at Ulu Burun (Ka#): 1984 Campaign.’ In American Journal of Archaeology. Vol.90, No.3, pp.269-296. Archaeological Institute of America. Accessed through JSTOR. 04.02.07.


Bass, G.F., 1987 ‘Oldest Known Shipwreck Reveals Splendors of the Bronze Age.’ In National Geographic. Vol.172, No.6, pp.693-733. The National Geographic Society.


Bewley, R., 2003 Prehistoric Settlements. Gloucestershire: Tempus Publishing.


Blandford, P., 1974 An illustrated history of Small Boats: A history of oared, poled and paddled craft. Buckinghamshire: Spurbooks.


Bradley, R., 1985 ‘Exchange and Social Distance- the structure of bronze artefact distributions.’ In Man. Vol.20, No.4, pp.692-704. Royal Anthropological Institute of Great Britain and Ireland. Accessed through JSTOR. 23.02.07.


Bradley, R., 1988 ‘Hoarding, Recycling and the Consumption of Prehistoric Metalwork: Technological Change in Western Europe.’ In World Archaeology. Vol.20, No.2, Hoards and Hoarding, pp.249-260. Taylor &amp: Francis Ltd.


Cleland, H.F., 1927 ‘Commerce and Trade Routes in Prehistoric Europe’. In Economic Geography, Vol.3, No.2, pp. 232-238. Accessed by JSTOR 15.02.07.


Collis, J.R., Gilbertson, D.D., Hayes, P.P., Samson, C.S., 1984 ‘Prehistoric and Medieval Field Archaeology of Crownhill Down, Dartmoor, England.’ In Journal of Field Archaeology. Vol.11, No.1, pp.1-12. Boston University. Accessed by JSTOR 25.02.07.


Craddock, P., and Lang, J. (eds) 2003 Mining and Metal Production Through the Ages. London: The British Museum Press.


Cunliffe, B., 1988 Mount Batten, Plymouth; A Prehistoric and Roman Port. Oxford: Oxford University Committee for Archaeology and individual authors.


Cunliffe, B., 2001 Facing the Ocean: the Atlantic and its Peoples. Oxford: Oxford University Press.


Cunliffe, B., 2002 The Extraordinary Voyage of Pytheas the Greek: The man who discovered Britain. London:Penguin Books.


Darvill, T., 2002 Prehistoric Britain. New York: Routledge.


Devon Archaeological Society, 2005 ‘Tin Peaks: Evidence for Roman and prehistoric tin mining on Dartmoor.’ In Devon Archaeological Society. No.90, pp.5.


Fleming, A., 1978 ‘The Prehistoric Landscape of Dartmoor, Part 1. South Dartmoor.’ In Proceedings of the Prehistoric Society. Vol.44, pp.97-123. Cambridge: Heffers Printers.


Fleming, A., 1988 The Dartmoor Reaves: Investigating Prehistoric Land Divisions. London: B.T. Batsford.


Fox, A., 1957 ‘Excavations on Dean Moor, in the Avon Valley, 1954-1956. The Late Bronze Age Settlement.’ In Transactions of the Devonshire Association for the Advancement of Science, Literature and Art. Vol.89, pp.18-77. Ministry of Works, Ancient Monuments Branch.


Fox, A., 1995 ‘Tin Ingots from Bigbury Bay, South Devon’. In Devon Archaeological Society Proceedings. No.53 pp. 11-23.


Gardiner, R., (ed) 2004 The Earliest Ships: The Evolution of Boats into Ships. London: Conway Maritime Press.


George, S.J, 1991 ‘Investigations In Bigbury Bay, Devon.’ In Nautical Archaeology Society Newsletter. No.4, pp.2-4.


Gerrard, S., 2000 The Early British Tin Industry. United Kingdom: Tempus Publishing.


Greeves, T., 1984 Archaeology of Dartmoor: an introduction. Dartmoor National Park Department.


Hauptmann, A., Maddin, R., and Prange M., 2002 ‘On the Structure and Composition of Copper and Tin Ingots Excavated from the Shipwreck of Uluburun.’ In Bulletin of the American Schools of Oriental Research. No.328, pp. 1-30.


Hobbs, R., Honeycombe, C., and Watkins, S., 2002 Guide to Conservation for Metal Detectorists. Gloucestershire: Tempus Publishing.


Lyons, C., Evans, A., Balfour, H., Smith, R.A., Peake, H., Warren, S.H., Evans, J.W., Crawford, O.G.S., 1912 ‘The Distribution of Early Bronze Age Settlements in Britain: Discussion.’ In The Geographical Journal. Vol.40, No.2, pp.198-203. The Royal Geographical Society. Accessed by JSTOR 15.02.07.


McDonald, K., 1993 ‘Devon’s Bronze Age tin.’ In Diver. Vol.38, No10, pp.26-28.


McGrail, S., 1998 Ancient boats in North-West Europe: The archaeology of water transport to AD 1500. New York: Addison Wesley Longman.


McGrail, S., 2004 ‘North-West seagoing boats before AD 400.’ In P. Clark (ed) The Dover Bronze Age Boat in Context. Society and water transport in prehistoric Europe. Pp. 51- 66. Oxford: Oxbow Books.


McLeish, A., 1992 Geological Science. Surrey: Thomas Nelson and Sons.


Muckelroy, K., 1980 ‘Two bronze age cargoes in British waters.’ In Antiquity. Vol. LIV, pp.100- 109. Antiquity Publications.


Muckelroy, K., 1981 ‘Middle Bronze Age Trade between Britain and Europe: a maritime perspective’. In Proceedings for the Prehistoric Society. Vol.47, pp.275-297. Cambridge: Heffers Printers.


Muckelroy, K, and Baker, P., 1979 ‘The Bronze Age site off Moor Sand, near Salcombe, Devon: An interim report on the 1978 season.’ In The International Journal of Nautical Archaeology. Vol.8, No.3, pp.189-210. The Nautical Archaeology Trust.


Muckelroy, K., and Baker, P., 1980 ‘The Bronze Age site off Moor Sand, Salcombe, Devon: An interim report on the 1979 season.’ In The International Journal of Nautical Archaeology. Vol.9, No.2, pp.155- 158. The Nautical Archaeology Trust.


Muhly, J.D., 1985 ‘Sources of Tin and the beginnings of Bronze Metallurgy’. In American Journal of Archaeology. Vol.89, No.2, pp.275-291. Archaeological Institute of America. Accessed through JSTOR. 15.02.07.


Muhly, J.D., Stech Wheeler, T., and Maddin, R., 1977 ‘The Cape Gelidonya Shipwreck and the Bronze Age Metals Trade in the Eastern Mediterranean.’ In Journal of Field Archaeology, Vol.4, No.3, pp.353-362.


Needham, S., 1980 ‘An assemblage of Late Bronze Age metalworking debris from Dainton, Devon.’ In Proceedings of the Prehistoric Society. Vol.46, pp.177-215.


Needham, S., 1988 ‘Selective Deposition in the British Early Bronze Age.’ In World Archaeology. Vol.20, No.2, Hoards and Hoarding, pp.229-248. Taylor &amp: Francis Ltd.


Needham, S., 2001 ‘When Expediency Brooches Ritual Intention- the Flow of Metal between Systemic and Buried Domains.’ In The Journal of the Royal Anthropological Institute. Vol.7, No.2, pp. 275- 298. Royal Anthropological Institute of Great Britain and Ireland. Accessed through JSTOR. 16.02.07.


Northover, J.P., 1982 ‘The exploration of the Long Distance Movement of Bronze in Bronze and Early Iron Age Europe.’ In Bulletin of Institute of Archaeology.’ Vol.19, pp.45-72.


Northover, J.P., 1988 ‘The Late Bronze Age metalwork: general discussion.’ In B. Cunliffe Mount Batten, Plymouth; A Prehistoric and Roman Port. Pp. 75- 85.Oxford: Oxford University Committee for Archaeology and individual authors.


O’Brien, W., 1996 Bronze Age Copper Mining in Britain and Ireland. Buckinghamshire: Shire Publications.


Palmer, D., and Easterbrook, G., (eds) 1999 S260 Geology. Block 2. Earth Materials. The Open University.


Parham, D., Needham, S., and Palmer, M., 2006  ‘Questioning the Wrecks of Time’ In British Archaeology. November/December 2006, pp.43-47.


Parker Pearson, M. 1993 Bronze Age Britain. London: B.T. Batsford Ltd.


Penhallurick, R.D. 1986 Tin in Antiquity. London: The Institute of Metals.


Pettit, P., 1974 Prehistoric Dartmoor. David and Charles (Holdings).


Price, D.G., 1993 ‘Dartmoor: the pattern of prehistoric settlement sites.’ In The Geographical Journal. Vol.159, No.3, pp.261-280. The Royal Geographical Society. Accessed through JSTOR. 16.02.07.


Pryor, F., 2004 ‘Some thoughts on boats as Bronze Age artefacts.’ In P. Clark (ed) The Dover Bronze Age Boat in Context. Society and water transport in prehistoric Europe. Pp. 31- 34. Oxford: Oxbow Books.


Reid, C., 1918 ‘Bronze and Tin in Cornwall.’ In Man. Vol.18, pp.9-11. Royal Anthropological Institute of Great Britain and Ireland. Accessed through JSTOR. 23.02.07.


Rickard, T.A., 1932 The Falmouth Ingot of Tin. In Man. Vol.32, pp.195-196. Royal Anthropological Institute of Great Britain and Ireland. Accessed through JSTOR. 23.02.07.


Roberts, O., 2004 ‘Round the headland or over the horizon?’ In P. Clark (ed) The Dover Bronze Age Boat in Context. Society and water transport in prehistoric Europe. Pp. 31- 34. Oxford: Oxbow Books.


Rohl, B., and Needham, S., 1998 ‘The Circulation of Metal in the British Bronze Age: The Application of Lead Isotope Analysis.’ In British Museum Occasional Paper. No.102. London: The British Museum.


Rowlands, M.J., 1971 ‘The Archaeological Interpretation of Prehistoric Metalworking.’ In World Archaeology. Vol.3, No.2 Archaeology and Ethnography, pp.210-224. Taylor &amp: Francis Ltd.


South West Maritime Archaeological Group (SWMAG), 2006 Report on the Erme Estuary Protected Wreck Sites 2005/2006 British Sub Aqua Jubilee Trust 2006. Privately Published Site Report.


Thibault, P.J., 1908 Metallurgy of Tin. Australia: Australian Mining Standard.


Tylecote, R.F., 1962 Metallurgy in Archaeology. Netherlands: A.W. Sijthoff’s Uitgeversmaatschappij N.V.


Tylecote, R.F., 1987 The Early History of Metallurgy in Europe. New   York: Longman Inc.


Tylecote, R.F., 1990 The Prehistory of Metallurgy in the British Isles. London: The Institute of Metals.


Tylecote, R.F., Photos, E., and Earl B., 1989 ‘The Composition of Tin Slags from the South-West of England.’ In World Archaeology. Vol.20, No.3, pp.434-445). Taylor & Francis Ltd.


Van de Noort, R., 2004 ‘The Humber, it’s sewn-plank boats, their contexts and the significance of it all.’ In P. Clark (ed) The Dover Bronze Age Boat in Context. Society and water transport in prehistoric Europe. Pp. 90- 98. Oxford: Oxbow Books.


Williamson, D., 1998 The mariners of ancient Wessex: A brief maritime history of central southern England up to the reign of Henry V111. Privately Published., 2007 Dover [online] Accessed on 27.04.07


Appendix A


British Bronze Age Dates with Relevant Metalworking Phases and related Boats



Years   BC

Metalwork   Phase


Final Neolithic

Early Bronze Age

Middle Bronze Age

Late Bronze Age

Earliest Iron Age

2500- 2200

2200- 2000

2000- 1900

1900- 1700

1700- 1500

1500- 1400

1400- 1300

1300- 1150

1150- 1000

1000- 800

800- 600



Mile cross



Acton park




Blackmoor/ Ewart

Llyn Fawr

Ferriby Boat 3

Ferriby Boats 1 & 2

Kilnsea Boat plank

Dover Boat

Shardlow logboat


Carpow logboat

Brigg ‘raft’


After Parham et al (2006, 47).


Appendix B


Table of Ingots


Ingot No


Size (cm)

Weight (kg)













































Plano- convex

Plano- convex

Plano- convex

Plano- convex

Plano- convex

Plano- convex

Plano- convex




Plano- convex

Plano- convex (hole in centre)

Plano- convex

Plano- convex

Plano- convex

Plano- convex

Plano- convex

Plano- convex

Plano- convex (broken)

Plano- convex

Plano- convex


Plano- convex


Plano- convex

Plano- convex




Plano- convex

Plano- convex

Plano- convex (broken)

Plano- convex



Sub- rectangular

Plano- convex

Plano- convex

Plano- convex






27.5 x 17 x 10

18.5 x 15.5 x 5

19.5 x 15 x 3.5

9 x 7 x 2.7

19.5 x 17.5 x 4.5

41 x 21 x 6.5

7.8 x 7.5 x 2

13 x 6 x 2

14 x 11 x 3

15 x 13 x 3

10 x 9 x 2

11 x 10 x 2

20 x 14 x 4.5

15 x 14 x 4

21 x 18 x 5

15 x 14 x 3

12 x 9 x 3.5

14 x 10 x 4

15 x 9 x 4

16 x 10 x 2.5

9.5 x 8.5 x 2

7 x 4

19 x 17 x 4

16 x 11.5 x 5

15.5 x 10 x 3

21 x 20 x 4

16 x 13 x 1


9 x 10 x 2.5

14 x 13 x 4

14 x 10 x 2

12 x 11 x 3

15 x 14



8 x 4 x 2.5

11 x 10 x –

11 x 9 x 2

11 x 7

13 x 10 x 3

22 x 16 x 3.5

10 x 6.5 x 3
















































This table shows the shape, size, and weight of each individual ingot lifted from the Erme Estuary Ingot Site. The variability between each ingot should be noted. The table does not include full information on every ingot and the measurements and weights were taken from Fox (1995, 17). The total weight of all ingots is 84.67 kg; however a percentage of this may be a result of impurities such as stoney material.


Appendix C


Plate 1. A selection of ingots from the Erme Ingot Site (After Fox 1995, 13).


Plate 2. Drawings of the Two H-Shaped Ingots by S. Rouillard (After Fox 1995, 18).




Plate 3. A selection of Ingots from the Erme Ingot Site drawn by S. Rouillard (After Fox 1995, 19).




Plate 4. A selection of Ingots from the Erme Ingot Site demonstrating various shapes (After Fox 1995, 20).




Plate 5. Ingot Number 6 showing high corrosion and a unique shape (After Fox 1995, 15).


Appendix D


Methodology for X-Raying the Tin Ingots


Once the x-ray machine is warmed up take the following steps.


1) Turn on room red lights and switch off the main lights at wall switch.


2) Load the appropriately sized film into cassette (ensure foil packaging and box lid are   re-secured).


3) Place items to be x-rayed (ingots) on top of the closed cassette containing film (clips facing down).


4) Place items inside x-ray chamber in centre of shelf.


5) Close the door firmly, taking care not to dislodge unstable samples. Door must be kept closed at all times except briefly during loading/unloading.


6) Set rotary timer to desired duration.


7) Press white warning buzzer button


8) Press blue start button


9) Set KVP to desired value immediately after pressing start.


10) Red light ‘x-ray’ will turn on.


11) Wait for red light to go out when the timer returns to zero.


12) Reset KVP control to zero manually.


13) Remove objects and film cassette from machine and close door firmly.


14) Unload film from cassette and develop.


Develop the film using developer (4:1 water) and fixer (4:1 water) for one minute in each and then run the film under tap water for 10minutes and allow to dry.


Times and Voltage (KVP) for each X-rayed Ingot of Tin


Ingot   Number

KVP   (Tube Voltage)

Time   (min)

Outcome/   Position
























Facing   up/ not enough   exposure

Facing   down/ not enough   time

Facing   down/ better

Facing   up/ Good


Developing   went wrong





Times and KVP varied dependant upon how good the exposure was. Longer time and higher KVP was used for larger ingots with a greater thickness.


The five ingots mentioned in the above table produced better results than those not listed.


Appendix E


Formation of crystals



Photograph demonstrating how experimental work can assist in learning more about the structure of ancient ingots (Tylecote 1987, 196).


Appendix F


Map to show the distribution of a number of Bronze Age sewn-plank boats in England (Van de Noort 2004, 93).