Archaeobotanical evidence for early date consumption in the Arabian Gulf

Abu Dhabi Authority for Culture and Heritage (ADACH)
P.O. Box 2380
Abu Dhabi
United Arab Emirates
email: mark.beech@cultural.org.ae

This web page is the online version of the following article:

Beech, M. 2003. Archaeobotanical Evidence for Early Date Consumption in the Arabian Gulf. Pages 11-31. In: ECSSR (ed), The Date Palm - From Traditional Resource to Green Wealth. Emirates Center for Strategic Studies and Research, Abu Dhabi. Paperback edition ISBN 9948-00-550-3.
[ you can read an on-line version by clicking here - read a summary of the contents of the book in english or in arabic, or if you wish to purchase the entire book click here ]

This was a paper presented at the International Date Palm Forum, which was held in Abu Dhabi, United Arab Emirates between the 15-17 September 2002. The International Date Palm Forum was organised by the Emirates Center for Strategic Studies and Research & the Ministry of Culture, Abu Dhabi, U.A.E.

It has long been known that the value of the date lies in its high nutritive value - the sugar content of dry dates can be 70% (Renfrew 1973). Archaeological evidence for the early exploitation of the date palm (Phoenix dactylifera) in the Arabian Gulf is provided by a number of different sources. These range from actual botanical remains in the form of charcoal (Tengberg 1998; Tengberg, this volume) to the survival of the date stone or whole dates (Beech and Shepherd 2001; Costantini 1985; Nesbitt 1993; Renfrew 1973; Rowley-Conwy 1987) to microscopic traces such as phytoliths. Date palm phytoliths have been successfully identified from a 1st century BC - 1st century AD layer near the main entrance of a temple at ed-Dur, Umm al-Qaiwain, U.A.E. (Haerinck et al 1998). This particular deposit, along with a bronze ring seal, illustrating a person holding what appears to be a palm leaf in their hand, clearly illustrates the symbolic as well as economic importance of dates in the region.

New biomolecular evidence may now also be used to examine pottery and other vessels for residual traces of date juice, as in the case of recent work carried out on material from the Nubian site of Qasr Ibrim in Egypt (Copley et al. 2001a, b). The chemical analysis of absorbed residues in archaeological pottery is well established, and through the investigation of ceramic vessels (via gas chromatography, gas chromatography-mass spectrometry and gas chromatography-combustion-isotope ratio mass spectrometry) saturated carboxylic acids in the range C12 to C18 have been detected (with an unusually high abundance of C12) from vessels from the Nubian site of Qasr Ibrim. This is mirrored in the saturated fatty acid distributions detected from the kernels of modern and ancient date palm (Phoenix dactylifera L.) and dom palm (Hyphaena thebaicas (L.) Mart.). Mixing in some of the vessels of the palm fruit with another lipid source is indicated through the C13 values. These results provide some of the first direct evidence for the use of pottery vessels in the processing of palm fruit.

The remains which archaeologists most often encounter, however, consist of either date palm charcoal or carbonised date stones. Such remains have been preserved rather fortuitously by charring. In certain circumstances date stones may be preserved not by charring but by a process of mineralization. This occurs when the local burial conditions are such that the high degree of calcium carbonate leads to mineral replacement and the date stones become literally fossilized. Such remains have a typical hard and stone like quality. Another source of evidence are mud brick and pottery fragments bearing impressions of date palm charcoal or date stones (Beech and Shepherd 2001; Rowley-Conwy 1987).

Finally, there is information available from various historical and art sources. My aim here, however, is not to discuss these latter forms of evidence, as these apply more to later periods from the Bronze age onwards when literary evidence in the form of cuneiform sources are available (cf. Potts, this volume).

The chronological focus for this present study examines the earliest evidence for the exploitation of the date palm in the Arabian Gulf, concentrating on the period from about 7,500 to 5,000 years ago (ca. 5500 - 3000 BC). This time period is broadly known as the Arabian Neolithic. A number of coastal sites in the region are characterized by the presence of Ubaid pottery, a type of pottery so named from the type site of Tell Al Ubaid, located in southern Mesopotamia, present day Iraq. This pottery which was traded from southern Iraq is found at a number of sites stretching down from Iraq to Saudi Arabia, Bahrain, Qatar and the United Arab Emirates (Oates et al. 1977; Carter and Crawford 2001; Shepherd Popescu and Beech, in press).

How are the date palm remains discovered on archaeological sites? Up until thirty years ago most archaeological excavations were only ever carried out by hand. This meant that the majority of small and especially delicate items were lost as they were simply missed by the excavator. The advent of dry and wet sieving, followed by the introduction of flotation methods to retrieve botanical material on excavations has substantially improved our knowledge and understanding of archaeobotanical remains in this region. Most excavations currently conducted in the region undertake some form of dry sieving during the excavation, and take bulk samples of sediment which are washed using a flotation system to retrieve fine botanical material. Each of these recovery methods of course has its own biases and limitations. Dry sieving may destroy more fragile remains. Wet sieving may be more gentle, although may be less practical if the excavation site is located a long way from the nearest water source. Using a flotation system may successfully retrieve the fine botanical fraction but may prove to be very labour intensive for not much gain, e.g. the archaeobotanist, Dr. Mark Nesbitt, from the Centre of Economic Botany at Kew Gardens, who worked on the 2nd mill. BC Dilmun settlement of Saar in Bahrain, had to process 6804 litres of sediment to obtain just 82g of botanical remains! (Nesbitt 1993: 24).

Preservation conditions also effect the survival of date palm remains. Date stones may be accidentally charred in the embers of a fire, but this can lead to some size changes, namely an overall decrease in size. An experiment where a sample of 24 modern date stones were carbonized, 12 at 150 degrees C and 12 at 240 degrees C, revealed that major changes in dimensions occurred during carbonization, and these were greater at the higher temperature; length, breadth and thickness all decreased (Rowley Conwy 1987: 183). Mineralised date stones may better represent the original size of the stone although measurements may slightly overestimate or exaggerate the actual size. Impressions of date stones in pottery and mud brick are likely to represent uncarbonised examples incorporated into the fabric during manufacture, so they may be closer to the original size of the date stone.

In this contribution I would like to focus on two broadly contemporary sites which both date to around 7,000 years ago. These are the sites of DA11 on Dalma Island located in the western region of Abu Dhabi in the United Arab Emirates, and the site of H3 at Sabiyah in NE Kuwait. Both these sites have certain similarities but also contrasts. Structures in the form of houses are present at both sites, as well as a wealth of other archaeological material. Ubaid pottery was recovered at both sites, although sites in United Arab Emirates have far less quantities of pottery when compared to their northern counterparts.

Dalma is located some 45km off the coast in the western region of Abu Dhabi (Figure 1). The site, called DA11, is located within Dalma town within the former compound of the Dalma Women's Association. Work carried out between 1992 and 1994 as part of the Abu Dhabi Islands Archaeological Survey established the presence of an early Neolithic beach settlement with structures and middens (Flavin and Shepherd 1994). Small quantities of imported painted 'Ubaid ware from southern Mesopotamia were recovered, along with a large assemblage of locally made gypsum plaster vessels (Joyner et al, in prep.). Some of these were painted with line and chevron patterns, apparently imitating the foreign imported Ubaid pottery (Shepherd Popescu and Beech, in press). Many thousands of flint flakes and numerous tools (including drills, arrowheads, scrapers and tile knives) were found, as well as nearly a hundred ornamental beads and pendants of varying type. Food debris took the form of marine mollusca and animal remains, including a substantial assemblage of fish bones. Domestic sheep or goat was present, and gazelle seem to have been occasionally hunted (Beech 1999, 2001; Beech and Glover in press). Further excavations by the Abu Dhabi Islands Archaeological Survey team in 1998 revealed further important traces of the settlement. This confirmed the presence of at least two round house-like structures with surviving post-holes and floors (Beech and Elders 1998, 1999).

A fragmentary carbonised date-stone was recovered during the 1994 season from a redeposited sand layer just below the present day ground surface (context 4). During the excavation in 1998 of a burnt layer or possible hearth (context 15, first identified in 1993), located about 25cm above the floor level of one of these house structures, several further interesting archaeobotanical finds were made. These were a complete carbonised date stone as well as two fragments of burnt mud brick, which had impressions of date stones within them (Figures 2 -3). Both these carbonised date stones were AMS radiocarbon dated by the Scottish Universities Research and Reactor Centre (SURRC) radiocarbon laboratory at the University of Glasgow, in conjunction with the University of Arizona AMS facility. The details of their findings are presented in Figure 4. Calibrations are made using the OxCal v.3.5 software program, using the datasets derived from Stuiver et al (1998). The decadal atmospheric calibration curve is used. Calibrated age ranges are calculated with 2 sigma errors from the probability distributions. The relative area under the probability distribution is given in brackets after the age range.

The radiocarbon dates confirmed that the inhabitants of Dalma were exploiting date palms at least 7,000 years ago. Details of these findings were subsequently published in the international archaeological journal Antiquity in 2001 (Beech and Shepherd 2001).

The previous earliest evidence for date palm (Phoenix dactylifera) remains in the United Arab Emirates were the date palm imprints excavated from Hili 8 (Period 1 - Building VI deposit), dating to around 3000 BC (Cleuziou and Costantini 1980). Date charcoal and stones also occur in late 3rd - early 1st millennium levels at Tell Abraq, U.A.E. (Potts 1990). Elsewhere in the Gulf, date-stones have been recovered from Failaka, Kuwait, dating to 2000 BC (Rowley-Conwy 1987) and from Qala't al-Bahrain, Bahrain, dating to 1475 BC (Potts 1990). This means that the Dalma dates are at least two thousand years older than the previously known findings from Hili in Al Ain.

Archaeobotanical records for dates in south-west Asia have been summarised by Nesbitt (1993). Little is known about the domestication of the date palm, partly because it has proved very difficult to distinguish truly wild palms from feral escapes. The original wild ancestors could have grown somewhere in North Africa, Arabia, in the southern parts of the Near East or in the Indus basin. Dates are cultivated in all these areas at the present day. According to Nesbitt (1993, 30), the only reliable archaeobotanical records of its occurrence on early sites are: Tepe Gaz Tavila, located near Daulatabad south of Kerman in southeastern Iran - 5400-4800 BC (Costantini 1985), and Tell el-Ouelli, Iraq - 5th mill. BC (Huot 1988). Costantini (1985: 214) has reported two uncarbonised, silicified date-stones from Mehrgarh in Pakistan, dating to 6000 and 5000 BC. As these are uncarbonised, however, their date and contextual provenance may be questionable. Zohary and Hopf (1988) mention the presence of a few date palm kernels from Egypt, Iran and Pakistan dating to the 6th and 5th millennia BC, but dismiss them as probably representing material collected from the wild. They say that the earliest remains of what seem to be cultivated dates are those excavated by Seton Lloyd in the Ubaid horizon (c.4000 BC) at Eridu in Lower Mesopotamia. Zohary and Hopf (1988: 150) suggest that the date palm was first brought into cultivation somewhere in the lower Mesopotamian basin, or in some oases in the southern fringe of the Near Eastern arc. Taking all the existing records as a whole, it seems likely that dates have been cultivated since at least the fifth millennium BC (Nesbitt 1993: 31; Zohary and Hopf 1988: 149).

In this context, the discovery and dating of the Dalma carbonised date-stones is therefore of some interest, particularly as their age range falls within the late 6th - early 5th millennia BC. They represent some of the earliest evidence for date consumption found within the Middle East. Although it cannot be determined whether they represent wild or cultivated dates, it is at least clear that they were consumed at that time. They may have been harvested locally on Dalma but it also quite possible that they may have been brought as trade goods into the settlement (cf. Oates et al. 1977).

Since the time of the initial publication of the Dalma dates, the author has been involved in an archaeological research project in Kuwait which has provided new evidence for the early exploitation of date palms. My aim here is to introduce these new findings.

The site of H3 was first discovered in the early 1990's by a geologist colleague of Dr. Fahad al Wohaibi, the former director of the National Museum of Kuwait. The Kuwaiti-British Archaeological Expedition to Sabiyah was formed in the spring of 1998 as the result of a visit to Kuwait by Dr. Harriet Crawford (Institute of Archaeology, University College London) in the autumn of 1997. This took place at the invitation of the former Director of the National Museum, Dr Fahad al Wohaibi. A number of possible projects were discussed and permission was then granted by the Secretary General of the National Council for Culture, Arts, and Letters for a preliminary survey by the Expedition, to be carried out jointly with a team from the National Museum of Kuwait, of the area known as as-Sabiyah at the north end of Kuwait bay. The area had already been the subject of a preliminary study by Dr. Fahad who had identified one well preserved site (H3) with both painted and plain Ubaid pottery on the surface. It was agreed that the team should undertake exploratory work at this site. Four field seasons in 1998, 1999, 2001 and 2002 subsequently took place (Carter and Crawford 2001), with a team of British archaeologists and specialists, working together with Kuwaiti colleagues from the National Museum and National Council for Culture, Arts and Letters (NCCAL). This work was generously sponsored principally by Kuwait Shell, with additional funds from NCCAL, as well as a host of other sponsors (cf. acknowledgments).

The site of H3 is located on the northern edge of the western end of Jazirat Dubaij, a 4km long red quartzite sandstone bedrock promontory which extends westwards from the Jal Az-Zor escarpment (Figure 5). This promontory rises some 2-4 metres above the surrounding area of desert sabkha. The modern day coastline of Kuwait Bay is located some 4km to the south of the Jal Az-Zor ridge, and the present day bay and adjacent area of Bubiyan Island is now affected by a build up of mud and sand, which is believed to have largely taken place since the Bronze age. Beach deposits recorded at similar datums to the location of H3 at Jazirat Dubaij have been noted in the neighbouring region of Al Bahra (Al-Asfour 1982) and Bubiyan Island (Al-Zamel 1983). These features have yielded radiocarbon dates of around 4000 BP. At the time when site H3 was occupied the actual coastline would have been much closer to the site, and it possible that a natural lagoon and harbour existed immediately behind the site to the north of the Jazirat Dubaij.

Excavations at H3 have revealed a fascinating archaeological sequence. The site initially appears to have been a temporary campsite where people stopped by the coast to fish, process their shellfish, and cook on their campfires. This may have taken place on a seasonal basis. Later a set of stone buildings, consisting of a series of chamber like structures, was built immediately on top of the shell midden which had accumulated on the former shoreline. These were then sometimes modified, chambers being rebuilt or sub-divided and re-used. A number of structurally and functionally discrete areas could be identified, supporting the hypothesis that the site was now more than just a temporary fisherman's camp.

The pottery from H3 largely belongs to the period known as Ubaid 2/3, falling within the second half of the 6th millennium BC (Carter and Crawford in press). A single radiocarbon date is available from the site, based on a sample of ash from a hearth (context 70), which is contemporary or earlier than the date of the excavated buildings. The sample (AA-42171 - GU-9301) was submitted to the Scottish Universities Research and Reactor Centre in Glasgow, and was counted at the University of Arizona AMS facility. Calibration was made using the OxCal v.3.5 software program, using the datasets derived from Stuiver et al (1998). The decadal atmospheric calibration curve is used. The calibrated age range is calculated with 2 sigma errors from the probability distributions, and the relative area under the probability distribution is given in brackets after the age range. This gave a calibrated date within the third quarter of the 6th millennium BC (Figure 6).

Archaeological excavations at H3 unearthed an interesting range of finds which included a complete model ceramic boat, a drilled pearl, thousands of shell beads, evidence for the manufacture of pearl oyster buttons, sequins and pendants, body ornaments such as labrets (?lip plugs) and ?ear stoppers, a rich diversity of lithic tools and material including flint arrowheads, chert tools, polished axes and obsidian imported from Yemen.

Some of the most important findings during the 2001 and 2002 excavations were slabs of bitumen bearing interesting impressions. The outside of a number of pieces were encrusted with barnacles, suggesting that they had endured prolonged exposure to the sea. On examining the inner surface of a number of these pieces, clear traces of impressions left by bundles of reeds could be seen, along with a series of deliberate holes where perhaps pegs or rope had been pushed through. Similar type fragments have been discovered at the Bronze age site of Ras al-Jinz on the Omani coast, which have been interpreted as early evidence for reed boats, the outer surface of which was caulked with bitumen to keep them watertight and seaworthy (Cleuziou and Tosi 2000). Here at the site of H3 in Sabiyah we now have the earliest direct evidence of boats in all Arabia, these fragments pre-dating the Omani discovery by a further 3,000 years! The marsh arabs of southern Iraq are of course well known to us through the accounts of Thesiger. It is remarkable that this tradition may go back as far as 7000 years ago in this area. If such boats were seaworthy and could be navigated down the Gulf, then it is clear that exchanges of goods (including perhaps dates) could easily be taking place between different regions of the Gulf, even at this early time.

So how did the inhabitants of site H3 interact with their environment? We know that they certainly spent a proportion of their time fishing judging from the traces of fishing equipment discovered at the site, along with the large quantities of fish bones discarded (Beech 2001). The occupants of H3 maintained domestic livestock. Bones from cattle and sheep or goat have been identified. Hunting for gazelle and fox also took place. Marine resources exploited for food included sea turtle, crabs and sea birds like cormorant. Fishes exploited included requiem sharks, sawfish, eaglerays, sea catfish, flatheads, groupers, jacks, scad, queenfish, emperors, seabream and tuna (Beech in press a,b).

What did the coastline look like at this time? One of the most dominant shellfish species occurring at both H3, which was also common at Dalma, is the gastropod species, Lunella coronata, known commonly as the turban shell. This species typically inhabits rocky intertidal pools, quite unlike the appearance of the modern day coastline immediately adjacent to the site, as discussed earlier. Other important shellfish present at the site included thorny oysters (Spondylus marisrubri), pearl oyster (mostly Pincatada radiata) and the strombus shell (Strombus decorus persicus). These species again are not found in any great quantity on the modern day coastline. Many fragments of these shells were used in the manufacture of jewellery at the site.

Excavations at H3 in 2001 and 2002 revealed that the inhabitants of H3 also clearly exploited date palms. A total of three mineralised date stones (Figure 7) were recovered from dump layers within chambers 1 (context 1029), 11 (context 1515) and ?18 (context 1208). These were hard and stone-like due to the process of mineralization. The dense amount of shell material within the deposits, along with occasional ashy deposits, must have enabled the preservation of such material.

These new date stones from Kuwait provide valuable additional information to augment the biometric analysis of early date stones already published (Beech and Shepherd 2001). The standard measurements usually taken on date stones are as follows: length, breadth (laterally) and thickness (measured at the mid-point in a dorso-ventral direction) (Figure 8).

A valuable opportunity arose in the summer of 2002 to examine a large amount of modern date stones collected by Phil Iddison, a member of the Al Ain Committee of the Emirates Natural History Group. He kindly donated this date reference collection to the Abu Dhabi Islands Archaeological Survey for scientific research. Phil collected a total of 752 date stones from 70 varieties of dates which were sold in Al Ain and Buraimi markets during the course of the past few years. These largely originated from date farms in the Al Ain and Buraimi region, although some of the dates were Saudi and Iraqi in origin. What particularly drew me towards undertaking measurements of modern date stones was to assess the variability in their size and shape. In the analysis of the Dilmun 2nd mill BC material from Saar in Bahrain, Mark Nesbitt commented upon the fact that the size of the Saar date stones was very similar to those of Failaka, but stated that:

"… once larger numbers of stones are available for measurement, clusters of different sizes conforming to varieties may become apparent" (Nesbitt 1993: 28)

I was intrigued by this comment and wanted to investigate firstly the variability of measurements of some modern date varieties, and then secondly, how the archaeological date stones compared with these data.

The best separation into distinctive groups was obtained by plotting the length of the modern date stones against their thickness (Figure 9). Although there were few outlier cases in the case of the Raziz and Sukari varieties. These outliers actually turn out to be the same variety but from different sources. Most of the Raziz dates were from Al Ain and Buraimi area, whereas the seven smaller examples came from the Al Saad date factory. In the case of the Sukari dates, the three outliers may prove to have been misidentified as Phil Iddison records that they are "?sukari".

Unfortunately in the case of archaeological material it is often not possible to measure an accurate length measurement, as much of the material may be damaged. By measuring width against thickness a far large sample can be usually compared. This is why a number of authors have adopted such an approach (Rowley-Conwy 1987; Nesbitt 1993; Beech and Shepherd 2001). Comparing the breadth against thickness in the modern samples generally demonstrates a fairly good separation again of the different varieties, with the exception again of the Raziz and Sukari varieties, which can be explained as stated earlier (Figure 10).

An attempt was made to then compile a catalogue of measurements of archaeological date stones to see how these compared (Figure 11). The material is compiled in this table with the oldest material at the top of the table, with later periods being represented down towards the bottom of the table. Note that the older date stones tend to be slightly longer or at the top end of the range of measurements from Bronze age material. If one plots the breadth against thickness of all the archaeological date stones, one can see that the older material tends to be also broader and thicker.

Comparing the size (width versus thickness) of the Dalma date-stones with those published measurements available from a number of other contemporary and later sites in the region, it can be seen that the two Dalma examples fall at the upper end of the size range (Figure 12). The older stone from context 15 falls just outside the upper range of the Failaka and Saar distributions, whilst the younger one from context 4 falls within the upper part of the distributions from these sites. Interestingly all the largest date-stones belong to the early period sites such as Mehrgarh and Tepe Gaz Tavila, which are the closest in age to the Dalma specimens. It should be noted though that as the Mehrgarh examples were not carbonised this difference may be due to the fact that they have not suffered from shrinkage as a result of the burning process.

Superimposing the archaeological data with all the modern breadth against thickness data reveals an interesting pattern (Figure 13). Note that the axis of the archaeological measurements is shifted slightly upwards from that of the modern material. This is curious, because if Rowley-Conwy 1987 is correct that charring generally leads to an overall decrease in length, breadth and width, this means that the archaeological material must be even slightly larger again than the modern dates. It would be interesting to compare further measurements of archaeological date stones from other areas and chronological periods to see how they compare with this patterning. There certainly appears to be some degree of variation in the measurements. I would welcome receiving further modern and archaeological data from any botanists or archaeobotanists with measurement data on date stones.

In conclusion, the date palm is of course important but it cannot be separated from the entire social, economic and environmental package which made up the lives of the people inhabiting the Arabian Gulf during the Neolithic period, some 7,500 to 5,000 years ago (ca. 5500 - 3000 BC). The everyday economic life of people at this time included cattle and sheep/goat husbandry, marine resource exploitation, as well as the exploitation of dates.

Whether such dates were already cultivated or simply represent people harvesting wild dates remains open to question. It certainly is a fascinating possibility that already 7,500 years ago in the Arabian Gulf different groups in society may have already been beginning to specialize in the exploitation of particular resources. Work carried out by the author on examining the question of the seasonal occupation of these coastal sites by studying fish otoliths, suggests that whilst some part of the population may have fished in the summertime on the coast (Beech 2001; Beech in press a,b), other groups may followed the traditional pattern of headed inland to the oases, or to the mountains in search of water and pasture for their animals (Uerpmann and Uerpmann 2000; Uerpmann et al. 2000).

(1) The earliest evidence for exploitation of Phoenix dactylifera in the Arabian Gulf can be found at the archaeological sites on Dalma, UAE and Sabiyah, Kuwait. These sites were both inhabited around 7000 years ago.

(2) It is not possible to say whether the remains represent wild or cultivated dates.

(3) Comparing the size of the archaeological date stones from Dalma and Sabiyah with other archaeological material as well as with modern date varieties from the UAE reveals the following trends:

archaeological (late 6th mill BC - Dalma & Sabiyah)
length range = 15.0-22.5mm, breadth range = 7.0-9.5mm, thickness range = 6.1-9.4mm

archaeological (3rd-2nd mill BC - Ur, Failaka, Saar and Hili 8)
length range = 10.4-19.5mm, breadth range = 4.6-7.5mm, thickness range = 4.8-6.8mm

modern varieties (UAE) ca 70 varieties, total sample size = 752
length range = 13.5-28.0mm, breadth range = 5.5-11.0mm, thickness range = 4.5-9.5mm

The older Neolithic dates tend to be slightly larger than their Bronze age counterparts, and at the upper end of the measurements of the modern varieties.

(5) Care should be taken in the interpretation of archaeological charred vs. mineralised remains because of the problems of size change. Perhaps further experimental work might profitably be done investigating different burning temperatures and burial conditions.

The author would like to thank the Emirates Center for Strategic Studies and Research (ECSSR) and the Ministry of Information and Culture of the United Arab Emirates for inviting him to participate in the International Date Palm Forum in Abu Dhabi.

Thanks go the following people for their assistance during this study: Phil Iddison (Emirates Natural History Group / Hyder Consulting, Al Ain, U.A.E.) - for donating his modern date stone collection to the Abu Dhabi Islands Archaeological Survey (ADIAS); Will Higgs (Department of Archaeology, University of York, U.K.) - for measuring and photographing the modern date stones; Dr. Mark Nesbitt and Dr. Sasha Barrow (Centre for Economic Botany, Royal Botanic Gardens, Kew) for sharing their thoughts and publications on date palms when I was doing the early phase of this work;

The field work on Dalma was carried out by the Abu Dhabi Islands Archaeological Survey (ADIAS), which was established in 1992 on the directives of His Highness Sheikh Zayed bin Sultan Al Nahyan, and operates under the patronage of His Highness Sheikh Mohammed bin Zayed Al Nahyan. The Project is charged with the responsibility of surveying for, recording, and, where necessary arranging for the excavation of archaeological sites on the coast and islands of the Western Region of Abu Dhabi. Its Academic Director is Dr. G.R.D. King, Reader in Islamic Art and Archaeology at the School of Oriental and African Studies, University of London, while the Executive Director in Abu Dhabi is Peter Hellyer. Excavations at the site were directed by Katelin Flavin and Elizabeth Shepherd during the 1993-4 seasons, and by Dr. Mark Beech and Dr. Joseph Elders during the 1998 season. The participation of Mark Beech in this work was carried out as a component of his PhD research in the Departments of Archaeology and Biology at the University of York. This research was financially supported by the University of York, ADIAS, the British Council (Abu Dhabi) and the Environmental Research and Wildlife Development Agency (ERWDA) in Abu Dhabi.

The Kuwaiti-British Archaeological Expedition to Sabiyah has been generously funded principally by Shell Kuwait, as well as by the National Museum of Kuwait and the National Council for Culture, Arts and Letters (NCCAL). Other funding bodies for the Kuwait project include the British School of Archaeology in Iraq; the Institute of Archaeology (University College London); the Charlotte Bonham-Carter Charitable Trust; the Central Research fund of London University; and the Society for Arabian Studies.

Al-Asfour, T. 1982. Changing Sea-Level Along the North Coast of Kuwait Bay. Keagan Paul International Ltd: London.

Al-Zamel, A. 1983. Geology and Oceanography of Recent Sediments of Jazirat Bubiyan and Ras As-Subiyah, Kuwait, Arabian Gulf. Unpublished Ph.D thesis. University of Sheffield.

Beech, M. 1999. Dalma archaeological site yields Arabia's oldest date stones. Tribulus (Bulletin of the Emirates Natural History Group) 9.1: 18.

Beech, M. 2000. Preliminary report on the faunal remains from an 'Ubaid settlement on Dalma island, United Arab Emirates. In: M. Mashkour, A.M. Choyke, H. Buitenhuis and F. Poplin (eds.), Archaeozoology of the Near East IV: Volume B - Proceedings of the fourth international symposium on the archaeozoology of southwestern Asia and adjacent areas. ARC Publicatie 32. Groningen, Netherlands. pp.68-78.

Beech, M. 2001. In the Land of the Ichthyophagi: Modelling fish exploitation in the Arabian Gulf and Gulf of Oman from the 5th millennium BC to the Late Islamic Period. DPhil thesis - Departments of Archaeology and Biology, University of York, U.K.

Beech, M. 2002. Fishing in the 'Ubaid: a Review of Fish-bone Assemblages from Early Prehistoric Coastal Settlements in the Arabian Gulf. Journal of Oman Studies 12: 19-34.

Beech, M. in press (a). The Development of Fishing in the United Arab Emirates: A Zooarchaeological Perspective. In: D.T. Potts (ed.), Proceedings of the First International Conference on Archaeology in the United Arab Emirates, Abu Dhabi, United Arab Emirates (15-19 April 2001). Ministry of Information and Culture, Abu Dhabi.

Beech, M. in press (b). Preliminary report on the vertebrate fauna from site H3, Sabiyah: an Arabian Neolithic/Ubaid site in Kuwait. In: L. Martin, H. Buitenhuis and A. Choyke (eds.), Archaeozoology of the Near East VI - Proceedings of the sixth international symposium on the archaeozoology of southwestern Asia and adjacent areas. ARC Publicatie. Groningen, Netherlands.

Beech, M. and J. Elders. 1998. UAE's oldest houses discovered. Tribulus (Bulletin of the Emirates Natural History Group) 8.1: 31.

Beech, M. and J. Elders. 1999. An 'Ubaid-related settlement on Dalma Island, United Arab Emirates. Bulletin of the Society for Arabian Studies 4: 17-21.

Beech, M., J. Elders and E. Shepherd. 2000. Reconsidering the 'Ubaid of the Southern Gulf: new results from excavations on Dalma Island, U.A.E. Proceedings of the Seminar for Arabian Studies 30: 41-47.

Beech, M. and E. Glover. In press. The environment and economy of an early 5th millennium BC site on Dalma island, United Arab Emirates. In: D. Baird and S. Campbell (eds.), Proceedings of the Fifth Millennium BC in the Near East Conference (Liverpool, UK). British Academy / British Institute of Archaeology at Ankara, British School of Archaeology in Iraq, and Council for British Research in the Levant.

Beech, M. and E. Shepherd. 2001. Archaeobotanical evidence for early date consumption on Dalma Island, United Arab Emirates. Antiquity 75: 83-9.

Carter, R. and H.E.W. Crawford. 2001. The Kuwait-British archaeological expedition to as-Sabiyah: report on the second season's work. Iraq 63: 1-20.

Carter, R. and H.E.W. Crawford. In press. The Kuwait-British archaeological expedition to as-Sabiyah: report on the third season's work. Iraq.

Cleuziou, S. and L. Costantini. 1980. Premiers elements sur l'agriculture protohistorique de l'Arabie Orientale, Paleorient 6: 245-51.

Cleuziou, S. and M. Tosi. 2000. Ra's al-Jinz and the prehistoric coastal cultures of the Ja'alan. Journal of Oman Studies 11: 19-73.

Copley, M.S., P.J. Rose; A. Clapham; D.N. Edwards; M.C. Horton and R.P. Evershed. 2001a. Processing palm fruits in the Nile Valley - biomolecular evidence from Qasr Ibrim. Antiquity, Sep 2001, Vol.75, No.289, pp.538-542.

Copley, M.S., P.J. Rose; A. Clapham; D.N. Edwards; M.C. Horton and R.P. Evershed. 2001b. Detection of palm fruit lipids in archaeological pottery from Qasr Ibrim, Egyptian Nubia. Proceedings: Biological Sciences, 22 March 2001, vol. 268, no. 1467, pp. 593-597(5).

Costantini, L. 1985. Considerazioni su alcuni reperti di palma da dattero e sul centro di origine e l'area du coltivazione della Phoenix dactylifera L., Pages 209-217. In: G. Gnoli & L. Lanciotti (eds.), Orientalia Josephi Tucci Memoriae Dicata: Rome: Istituto Italiano per il Medio ed Esremo Oriente, Serie Orientale Roma 56 (1).

Flavin, K. and E. Shepherd. 1994. Fishing in the Gulf: Preliminary investigations at an Ubaid site, Dalma (U.A.E.). Proceedings of the Seminar for Arabian Studies 24: 115-134.

Haerinck, E., L. Vrydaghs, and H. Doutrelepont. 1998. Des feux sacrificiels pour la divinité solaire à ed-Dur, Arabian Archaeology and Epigraphy 9: 125-130.

Huot, J-L. 1988. Ubaidian villages of lower Mesopotamia. Pages 19-42. In: E.F. Henrickson and I. Thuesen (eds.), Upon this Foundation - the 'Ubaid Reconsidered. Copenhagen: Museum Tuscalaneum.

Joyner,L., M. Beech and E. Shepherd Popescu. In prep. Ubaid plaster technology at Dalma island, UAE: raw materials, manufacture and decoration. Journal of Archaeological Science.

King, G.R.D. 1998. Abu Dhabi Islands Archaeological Survey: Season I - An Archaeological Survey of Sir Bani Yas, Dalma and Marawah: Season One. Trident Press: London.

Nesbitt, M. 1993. Archaeobotanical evidence for early Dilmun diet at Saar, Bahrain, Arabian Archaeology and Epigraphy 4: 20-47.

Oates, J.; T.E. Davidson; D. Kamilli and H. Mckerrell 1977. Seafaring Merchants of Ur?, Antiquity 51: 221-234.

Potts, D.T. 1990. The Arabian Gulf in Antiquity. Vol. I - from prehistory to the fall of the Achaemenid Empire. Oxford: Clarendon Press.

Renfrew, J. 1973. Fruits from Ancient Iraq: The Palaeoethnobotanical evidence. Bulletin on Sumerian Agriculture 3: 157-161.

Rowley-Conwy, P. 1987. Remains of date (Phoenix dactylifera) from Failaka, Kuwait. Pages 181-183, In: F. Hojlund (ed.), Danish archaeological investigations on Failaka, Kuwait. The second millennium settlements. Vol.2. The Bronze Age pottery: Aarhus: JASP, XVII(2).

Shepherd Popescu, E. and M. Beech (eds.). In press. Excavations at an early 5th millennium BC settlement on Dalma island, United Arab Emirates. British Archaeological Reports (International Series).

Stuiver, M.; P.J. Reimer; E. Bard; J.W. Beck; G.S. Burr; K.A. Hughen; B. Kromer; F.G. McCormac; V.D. Plicht and M. Spurk. 1998. Calibration issue, Radiocarbon 40: 1041-1083.

Tengberg, M. 1998. Paléoenvironnements et économie végétale en milieu aride: Recherches archéobotaniques dans la
région du Golfe arabo-persique et dans le Makran pakistanais (4ème millénaire av. notre ère – 1er millénaire de
notre ère), Univ. of Montpellier II, France. Unpublished PhD thesis.

Uerpmann, M. and H-P. Uerpmann. (2000). Faunal remains of Al-Buhais 18: An Aceramic Neolithic site in the Emirate of Sharjah (SE-Arabia) - excavations 1995-1998. In: M. Mashkour, A.M. Choyke, H. Buitenhuis and F. Poplin (eds.), Archaeozoology of the Near East IV: Volume B - Proceedings of the fourth international symposium on the archaeozoology of southwestern Asia and adjacent areas. ARC Publicatie 32. Groningen, Netherlands. pp.40-49.

Uerpmann, M., H-P. Uerpmann and S.A. Jasim. (2000). Stone age nomadism in SE-Arabia - palaeo-economic considerations on the Neolithic site of Al-Buhais 18 in the Emirate of Sharjah, U.A.E. Proceedings of the Seminar for Arabian Studies 30: 229-234.

Zohary, D. and Hopf, M. 1988. Domestication of Plants in the Old World. Oxford: Clarendon.

Length

Breadth

Thickness

FIGURE 8. Standard measurements taken on date stones.

FIGURE 9. Biometry of a sample of modern stones: length vs. thickness in mm.

FIGURE 10. Biometry of a sample of modern stones: breadth vs. thickness in mm.

COUNTRY	SITE	CONTEXT	MATERIAL	DATE	SAMPLE SIZE (n)	LENGTH (mm)	WIDTH (mm)	THICKNESS (mm)
Pakistan	Mehrgarh IB		mineralised date stone	6000 BC	1	20.7	7.7	6.9
Pakistan	Mehrgarh IIB		mineralised date stone	5000 BC	1	20.4	7.5	7.2
Kuwait	Sabiyah, site H3	A (west), context 1208	mineralised date stone	late 6th mill BC	1	19.7	8.8	8.3
Kuwait	Sabiyah, site H3	A (west), context 1029	mineralised date stone	late 6th mill BC	1	22.5	9.5	9.4
Kuwait	Sabiyah, site H3	A (east), context 1515	mineralised date stone	late 6th mill BC	1	15.0	7.7	6.4
UAE	Dalma, site DA11	context 4	charred date stone	4670 +/- 130 cal BC	1		7.1	6.1
UAE	Dalma, site DA11	context 15	charred date stone	5120 +/- 170 cal BC	1	20.0	7.6	6.4
UAE	Dalma, site DA11	context 15	mud brick impression	late 6th/early 5th mill BC	1	19.0	7.0
UAE	Dalma, site DA11	context 15	mud brick impression	late 6th/early 5th mill BC	1		7.0
Iran	Tepe Gaz Tavila		charred date stone	5400-4800 BC	1	18.0	7.5	6.8
UAE	Hili 8	IB	charred date stone	3000 BC	5	12.7 - 14.0	6.2 - 7.2	5.9 - 6.3
UAE	Hili 8	IIC	charred date stone	2600-2500 BC	15	11.1 - 16.2	5.1 - 6.8	4.8 - 6.7
Iraq	Ur	Queen Pu-abi's grave	charred date stone	mid 3rd mill BC	1	14.1	7.2	6.7
Iraq	Ur	Queen Pu-abi's grave	charred date stone	mid 3rd mill BC	1		5	4.9
Iraq	Ur	Queen Pu-abi's grave	charred date stone	mid 3rd mill BC	1		5.1	5.9
Iraq	Ur	Queen Pu-abi's grave	charred date stone	mid 3rd mill BC	1		5.1	4.8
Bahrain	Saar	1990: K17; 1991: E16, Q20, R20	charred date stone	1900 BC	16	10.4 - 19.5	4.6 - 7.2	5.0 - 6.3
Kuwait	Failaka	1129.AQS	charred date stone	1800 BC	4	14.8 - 18.7	5.7 - 7.5	4.8 - 6.8

FIGURE 11. Catalogue of measurements of archaeological date stones.
Data source: Failaka, Kuwait - 1800 BC (Rowley-Conwy 1987); Hili 8 IB, U.A.E. - 3000 BC (Costantini 1985); Hili 8 IIC, U.A.E. - 2600-2500 BC (Costantini 1985); Mehrgarh IB *, Pakistan - 6000 BC (Costantini 1985); Mehrgarh IIB *, Pakistan - 5000 BC Costantini 1985); Saar, Bahrain - 1900 BC (Nesbitt 1993); and Tepe Gaz Tavila, Iran - 5400-4800 BC (Costantini 1985). (* N.B.: the Mehrgarh examples are not carbonised).

FIGURE 12. Biometry of archaeological date stones: breadth vs. thickness in mm.

FIGURE 13. Comparison of biometry of archaeological and modern date stones: breadth vs. thickness in mm.