I recently helped organise the Integrated Microscopy Approaches in Archaeobotany workshop at the University of Reading, aiming to bring together researchers studying plant macrofossils, palynology, charcoal, waterlogged wood, phytoliths and anything else planty. Something that came out of the discussion at the end of the day was the need for a database of seed reference collections available for use. Whilst a Historic England funded zooarchaeological reference collection database is currently in the pipeline, with the potential for future archaeobotanical spin offs, it seemed sensible to quickly pull together a list of available reference collections in Britain and beyond. Other than Historic England, I’m afraid I don’t know how accessible these collections are.
Some other useful resources are the minutes of a 2010 AWG meeting and a very detailed article by Mark Nesbitt in Circaea 1991 discussing how to build a reference collection.
Historic England, Fort Cumberland
Reference collection of 4500 seeds and fruits, mainly British specimens. Also wood, charcoal, mosses and fibres.
“Researchers can visit the collections by prior arrangement with Ruth Pelling. We also lend out most of our accessions. Please note that students require prior approval from their supervisor(s) prior to using the collection and close supervision is rarely possible.”
This is just a quick post about an interesting new archaeobotany article by Erica Rowan. We did our DPhils together at Oxford, and Erica is now doing great research into Greek and Roman foods at Exeter. Carbonised olive stones are found quite often at archaeological sites in the Mediterranean. Rowan argues that much of this waste comes from the burning of olive pressing waste or ‘pomace’ as fuel. Pomace consists of olive flesh, stones, skin and seeds which are left over after olives are pressed for olive, and can be used for fertilizer, fodder or fuel. Pomace is still used as fuel today.
The tricky thing is how do we know that charred olive stones are evidence for the use of pomace as fuel, as opposed to burnt food waste or the remains of ritualised offerings. Rowan argues that the key factors are quantity and context. So, large quantities of charred olive stones in individual contexts or areas of an archaeological site. An example of a high density find is 100 items per L of soil floated for plant remains from the Cardo V sewer in Herculaneum. A high degree of fragmentation is also evidence for fuel use – more fragments of olive stones than intact pieces. Contexts which can indicate the use as pomace as fuel are given as places near kilns or domestic dwellings.
After establishing these criteria, Rowan goes on to summarise the current evidence for pomace use in the pre-Roman, Roman and Late Antique Mediterranean. Unsurprisingly, the Roman period is highlighted as one where archaeobotanical evidence for pomace fuel is found more often, especially in industrial and urban contexts. The identification of pomace as a common Roman fuel has big implications for modelling the fuel requirements of the Roman economy.
It would have been great to have some clearer recommendations about how to identify pomace use archaeobotanically, such as the number of samples needed within an archaeological site phase to reliably identify pomace use and the broad ranges of densities which indicate evidence of pomace as opposed to table waste and/or ritualised deposits. Either way, this paper is a great example of the more detailed approaches recently taken to the taphonomy of plant remains, be it digestive waste (O’Meara 2014), crop-processing (Antolin 2011), ritualised deposition (Lodwick 2015) and now fuel. Perhaps more importantly, it shows that archaeobotanical data can make major contributions to understanding something as heavily debated as the Roman economy.
Antolín, F., & Buxó, R. 2010. Proposal for the systematic description and taphonomic study of carbonized cereal grain assemblages: a case study of an early Neolithic funerary context in the cave of Can Sadurní (Begues, Barcelona province, Spain). Vegetation History and Archaeobotany, 20(1), 53–66. doi:10.1007/s00334-010-0255-1
Lodwick, L. (2015). Identifying ritual deposition of plant remains: a case study of stone pine cones in Roman Britain. In T. Brindle, M. Allen, E. Durham, & A. Smith (Eds.), TRAC 2014: Proceedings of the Twenty-Fourth Annual Theoretical Roman Archaeology Conference (pp. 54–69). Oxford: Oxbow.
O’Meara, D. (2014). Ruminating on the Past: A history of digestive taphonomy in experimental archaeology. In J. Reeves Flores & R. Paardekooper (Eds.), Experiments past: histories of experimental archaeology (pp. 131–145). Leiden: Sidestone Press. Link
Over the last few years I’ve been trying to establish what kind of arable farming was practiced at Late Iron Age Silchester. This is important as it in informs us about the people living in the Late Iron Age oppidum (were they farmers or craftsmen?), and how the community was organised (did people work together on the fields every day, or did they keep themselves to themselves?). The agricultural basis of Late Iron Age Silchester is also important for understanding the social processes behind the foundation of the settlement – did farmers from the surrounding region decide to move into a new area? Did trade with the continent stimulate the development of oppida, and were the residents having to generate grain surpluses to trade for imported dining wares and amphora? Or were these settlements focussed on pastoral activities, with very little arable farming taking place?
Much of what has been written previously on the agricultural basis of oppida in Britain has been based on conjecture, as few of these sites have been sampled for animal bones and plant remains when excavated *. The modern day landscape in which oppida are situated has heavily influenced opinions on the type of farming, or lack of, which was practiced at oppida. Most of these sites are located in areas of tertiary geological formations (usually clays, sands and silts), or where these intersect with calcareous areas or river gravels which are considered to have been settled earlier in prehistory. Regarding Silchester, Barry Cunliffe has written that “its own immediate hinterland may not have been particularly fertile, but it was able to thrive on the productivity of others.” (Cunliffe 2012, p 19). Niall Sharples has similarly written “The environment at all of these sites (including Silchester) is ill suited to arable agriculture as they are all … surrounded by nutrient-poor soils that support heath or woodland. Silchester also produced very little cereal-processing debris (chaff) despite extensive wet-sieving* (Fulford in Fulford and Timby 2000: 555) and it is suggested that this might indicate the site was dependent on other agricultural settlements” (Sharples 2010: 163-164).
George Boon, who synthesised the results of the Society of Antiquaries excavations at Silchester, had a more positive take on the potential of the land around Silchester. Writing in the 1970s, he said “The varied terrain around the town, therefore, offered the farmer light; well-drained but poor gravel soils to the west and north-west, for cultivation or for pasturage; and fertile arable slopes to the south, south-east and east” (Boon 1974, p 245). He did though link the limited amount of good arable land around Silchester with “the modest extent of built-up area of Calleva (Boon 1974, p 54).
So what is going on in the surroundings of Silchester today?
Late Iron Age Silchester is situated on the edge of a gravel plateau, stretching to the north and west. These soils are slightly acidic, free- draining and stony, so not ideal for arable farming. Today, there are extensive areas of conifer plantations, visible on satellite photos. These were established in the last century, to make the most of this area of land. There are also surviving areas of heathland, at Silchester Common and Tadley Common, which are an SSSI along with Pamber Heath. These are Calluna vulgaris (heather) – Ulex minor (gorse) heaths. Samples studied for waterlogged plant remains from the Late Iron Age forum-basilica (Jones 2000) and Insula IX (Lodwick 2014) show that bracken, gorse and heather were available in the surrounding area, but for how long heathland had been present is unclear.
To the south and east of Silchester are various damp areas of ground, as water collects at the intersection of the gravel and the underlying clay. An area of hay meadow is present at Ron Ward’s Meadow, another SSSI, alongside Honeymill Brook in Tadley. An area of fritillary meadow also survives to the north of Stratfield Saye (Brewis et al 1996, p. 43), the estate of the Duke of Wellington, although the valley was drained in the late eighteenth century as part of agricultural improvements (Curtler 1912, p505). It is likely that much more of this species diverse grassland used to be present in the region, as recalled in the areas of Tadley and Pamber (Tadley and District History Society 2013).
To the north-east of Silchester, in between the conifer plantations and heathlands of the gravel plateau and the hay meadows and pasture of the river valleys, are some areas of good arable land. The tertiary geological formations are very variable, so some of the areas are not too clayey and hard to cultivate, and not too sandy and free draining.
The area around Late Iron Age Silchester was much more variable than the broad notions of infertile marginal land allow for. Of course the only real way to establish the type of farming practiced at Silchester is to study the bioarchaeological remains from the site. I’ll be writing about this in future blogs.
* The Late Iron Age settlement at Stanwick, N Yorkshire, has been extensively sampled. Some archaeobotanical results are published in Van der Veen 1992, and the full report will be published soon in the monograph.
** True, there was not very much grain or chaff in Jones’ samples, but only 3L samples were taken (less than 10% of recommended sample size) and the published results are unclear
Boon, G. 1974. Silchester: the Roman Town of Calleva. Newton Abbot: David & Charles.
Brewis, A., Bowman, P., & Rose, F. 1996. The Flora of Hampshire. Colchester: Harley Books.
Curtler, W.H.R. 1912. Agriculture. In W. Page ed. The Victoria History of Hampshire and the Isle of Wight. Volume V. London: Constable and Company Limited,
Cunliffe, B. 2012. Calleva in context. In M. Fulford (Ed.), Silchester and the Study of Romano-British Urbanism (pp. 15–21). Portsmouth, Rhode Island: Journal of Roman Archaeology Supplementary Series Number 90.
Fulford, M., & Timby, J. 2000. Late Iron Age and Roman Silchester: Excavations on the Site of the Forum Basilica, 1977, 1980-86. London: Society for the Promotion of Roman Studies. Britannia monograph Series No. 15.
Jones, M. 2000. The plant remains. In M. Fulford & J. Timby (Eds.), Late Iron Age and Roman Silchester: Excavations on the Site of the Forum Basilica, 1977, 1980-86 (pp. 505–512). London: Society for the Promotion of Roman Studies. Britannia Monograph Series No. 15.
Lodwick, L. 2014. Condiments before Claudius: new plant foods at the Late Iron Age oppidum at Silchester, UK. Vegetation History and Archaeobotany 23: 543-549.
Sharples, N. 2010. Social Relations in Later Prehistory: Wessex in the First Millennium BC. Oxford: Oxford University Press.
Van der Veen, M. 1992. Crop Husbandry Regimes: an Archaeobotanical Study of Farming in Northern England 1000 BC – AD 500. Sheffield: J.R. Collis Publications.
Corncockle (Agrostemma githago) has been hitting the news in recent weeks, as Countryfile has been accused by The Telegraph for promoting the distribution of poisonous seeds via Kew Garden’s Grow Wild project. The over reaction has been called out by the Guardian online, and in a great blog post by Miles King. In short, corncockle does contain toxic compounds, but they are unlikely to cause ill health unless large quantities have been eaten.
The reason why corncockle seeds are being distributed by the Grow Wild project is that the plant is now very rare in Britain. The seeds rely on disturbed soil conditions to germinate, and are only distributed by being sown alongside cereal seed corn (Hellmund 2008).
Aside from the health risk aspect, corncockle also features in the complex debate on invasive alien species, as corncockle is an archaeophyte, having been introduced to Britain in the later first millennium BC. From an archaeological perspective, the identification of non-native species in archaeobotanical samples containing cereals can be used as evidence for the trade in cereal grain between areas where an arable weed is present, and where it is absent. in the later first millennium BC, corncockle is present in France and Germany (Hellmund 2008), but still absent from Britain and the Netherlands. Hence, it’s presence in southern Britain in the mid first century AD has been associated with cross-channel grain movements (Fuford 2004), related to the trade between oppida (Late Iron Age proto-urban settlements) and the Roman world, or the movements or supply of the Roman military.
Remains of corncockle are identified in several forms from archaeological sites. Most commonly, charred intact seeds are identified. The seeds are large triangular shapes, and have a distinctive papillose surface texture – see this image in the Netherlands seed atlas. The triangular tips of the seed capsule are also occasionally found in charred samples as well. Both the seeds and capsule fragments are of a similar size and weight to cereal grains, so that they both make it through the various crop-processing stages up until the hand-cleaning of cereal grains. Fragments of corncockle seeds are commonly found alongside cereal bran, chaff, grassland plants and bedding material, such as heather and bracken, in waterlogged conditions. Such an assemblage is commonly interpreted as material collected from stable floors, disposed of in pits or wells. The combination of fragmented seeds and cereal bran indicates that the material that has been fed to, and digested by, animals (probably horses) (Kenward and Hall 2012). Finally, fragments of corncockle seeds can also be found in calcium-phosphate mineralised deposits in latrine pits, such as at Saxon Winchester. Whole corncockle seeds were identified, as well as impressions of the seed coat (see Carruthers 2011 for pictures).
There hasn’t been an update of early archaeological corncockle records from Britain in a few years, whilst the date of introduction is usually given as during the later first millennium BC. So here’s a quick run down of some of the earliest prehistoric records in central-southern Britain, drawing from my Phd dataset:
Bury Hill, Hampshire. 1 charred corncockle seed was found alongside barley grain and spelt chaff from a Mid Iron Age pit, within a hill fort on the Hampshire downs (Campbell 2000).
Old Kempshott Lane. 1 corncockle seed was found alongside charred spelt grains and arable weeds from a pit in the Mid/Late Iron Age enclosed settlement, at the edge of the Hampshire Downs at Basingstoke (Haslam 2012)
Silchester, Hampshire. Corncockle seeds have been identified from waterlogged sediments at the oppida. First, in a Late Iron Age well at the Forum-Basilica site c. 25-15 BC (Jones 2000), and second from a later Late Iron Age well in Insula IX c. AD 30-55 (Lodwick 2014).
Suddern Farm, Hampshire. 5 seeds have been found in a charred deposit of barley and spelt crop-processing waste, from a pit in a Late Iron Age enclosed rural settlement (Campbell 2000).
Thruxton, Hampshire. 1 charred corncockle seed was identified from a Late Iron Age enclosure ditch, amongst spelt crop-processing waste (Campbell 2000).
Winchester, Northgate House. A few charred corncockle seeds were present alongside some barley and spelt wheat grains in a silty soil overlying Iron Age occupation (Carruthers 2011).
Following the Roman invasion in AD43, corncockle seeds become more common. Yet, as corncockle was already present in some areas pre-conquest, its identification cannot always be interpreted as indicating imported cereal grain from the continent. However, the weed species present amongst a few finds of cereal grain stores do indicate the long distance transport of cereals.
Isca grain, Caerleon. A deposit of charred grain from the Roman legionary fortress contained Agrostemma githago, Lathyrus cicerea, Raphanus raphanistrum and Vicia sativa (Helbaek 1964).
London forum grain. A layer of charred cereal grain has been studied from buildings lining the London Forum, destroyed by Boudica c. AD 61. The store was dominated by spelt wheat, and contained Agrostemma githago, Lens culinaris and Vicia ervilia.
Coney Street, York. A charred deposit of spelt wheat grains from a late first/early second century military granary contained Agrostemma githago, Consolida spp., Delphinium sp., Lathyrus aphaca and Raphanus raphanistrum (Williams 1979).
In these examples, the co-occurence of corncockle alongside other non-native weed seeds does suggest that cereals were imported from the continent.The presence of corncockle in later archaeobotanical studies, and the potential effects on health, are summarised here by Allan Hall. Whilst the focus in the last few weeks has been on the (over exaggerated) negative impacts of corncockle, the species is a very useful tool in archaeology!
Campbell, G. (2000). Plant utilization: the evidence from charred plant remains. In B. Cunliffe (Ed.), The Danebury Environs Programme. The Prehistory of a Wessex Landscape (pp. 45–59). Oxford: Institute of Archaeology.
Carruthers, W. (2011). Charred and mineralised plant remains. In B. Ford, S. Teague, & E. Biddulph (Eds.), Winchester – a City in the Making: Archaeological Excavations Between 2002 and 2007 on the Sites of Northgate House, Staple Gardens and the Former Winchester Library, Jewry St (pp. 363–373). Oxford: Oxford Archaeology.
Fulford, M. (2004). Economic structures. In M. Todd (Ed.), A Companion to Roman Britain (pp. 309–327). Oxford: Blackwell.
Haslam, R. (2012). Iron Age and Roman settlement and burial activity at Old Kempshott Lane, Basingstoke. Proc Hampshire Field Club & Archaeological Society 67(1): 79–141.
Helbaek, H. (1964). The Isca grain, a Roman plant introduction in Britain. New Phytologist 63(2): 158–164.
Hellmund, M. (2008). The Neolithic records of Onopordum acanthium, Agrostemma githago, Adonis cf. aestivalis and Claviceps purpurea. Vegetation History and Archaeobotany 17: 123–130.
Jones, M. (2000). The plant remains. In M. Fulford & J. Timby (Eds.), Late Iron Age and Roman Silchester: Excavations on the Site of the Forum Basilica, 1977, 1980-86 (pp. 505–512). London: Society for the Promotion of Roman Studies. Britannia Monograph Series No. 15.
Kenward, H., & Hall, A. (2012). Dung and stable manure on waterlogged archaeological occupation sites: some ruminations on the evidence from plant and invertebrate remains. In R. Jones (Ed.), Manure Matters: Historical, Archaeological and Ethnographic Perspectives (pp. 79–95). Farnham: Ashgate Publishing Limited.
Lodwick, L. (2014). Condiments before Claudius: new plant foods at the Late Iron Age oppidum at Silchester, UK. Vegetation History and Archaeobotany 23: 543-549.
Straker, V. (1984). First and second century carbonised grain from Roman London. In W. Van Zeist & W. Casparie (Eds.), Plants and Ancient Man: Studies in Palaeoethnobotany (pp. 323–329). Rotterdam: A. A. Balkema.
Williams, D. (1979). The plant remains. In H. K. Kenward & D. Williams (Eds.), Biological Evidence from the Roman Warehouses in Coney Street (pp. 52–62). York: Council for British Archaeology.
A couple of weeks ago, BBC Radio 4s Gardeners’ Question Time mentioned that Roman soldiers introduced Roman nettle to Britain. They used it to keep themselves warm by beating themselves with the stinging plants. As someone who studies Roman plant remains, this is a pretty amusing picture. But I was also a bit confused why I had never come across this in any of the archaeological material I read during my research.
First things first, what is Urticapilulifera? A quick flick through Clive Staces’ New flora of the British Isles (1997) shows Roman nettle used to occur as a casual in southern Britain, but does not occur any more. Next stop, the Online Atlas of the British Flora. No mention of Roman nettle, neither is there in Preston et al’s 2004 article on archaeophytes in Britain. Flora Europaea does tell us Roman nettle is an annual growing 30-100cm, monoecious, with 2-7cm leaves and flowers in globose heads. It now grows in southern Europe, but used to be found further north but is now rare.
Is this earlier distribution due to the expansion of the Roman Empire? The wikipedia page for Roman Britain has a brief section on environmental changes, in which it reiterates that roman soldiers introduced Roman nettle to warm themselves. Sadly, all of the fascinating plant foods that were introduced in the Roman period are missing (note:archaeobotanists should spend more time editing wikipedia). The article references a 1949 paper by Homer Nearing, published in the (paywalled) journal of the Medieval Academy of America. In this article Homer writes about the various medieval legends relating to Julius Caesar.
Homer discusses how the nettle story was described by William Camden as originating with John Parkinson, the Elizabethan herbalist. In his historical survey of Great Britain, (Britannia 1586), Camden describes how Parkinson linked Urticaromana with Romney, Kent by arguing that Julius Caesar landed there in 54 BC. The Roman soldiers apparently bought seeds with them to sow, so they could use the plants to keep warm. Camden disagrees with Parkinson on 2 points. Firstly, Roman nettle grew in several coastal locations in the late 16th century, and not just Romney. Secondly, Caesar didn’t land in Romney, so the story can’t be true. If Camden knew that the story was false over 400 years ago, how is it still accepted?
The association of stinging nettles with soldiers appears on various websites concerned with folk lore and natural remedies. Some websites associate the claim with William Camden, and miss off the bit where Camden disagrees with it. Here’s a good example. The association of Roman nettles with Roman soldiers also makes it way into medical journals, appearing in Randall et al’s 2000 paper on using nettle for pain relief. In turn, they reference Richard Mabey’s Flora Britannica, which attributes the link between Roman soldiers and Roman nettle back to Camden (Mabey 1996, p. 68).
There are currently no records of Urtica pilulifera listed on the Archaeobotanical Computer Database for Britain. Considering the hundreds of Urtica dioica – stinging nettle and Urtica urens – common nettle seeds that archaeobotanists find from archaeological settlements, the absence of Roman nettle must mean it was never introduced here.
Nettles are interesting for their use in Roman medicine, and Urtica urens (small nettle) was introduced to Britain in the Late Bronze Age (Preston et al. 2004). But the introduction of Roman nettle to Britain by roman soldiers is only a popular folk story.
As an archaeobotanist, I study plant remains from archaeological sites, usually seeds, cereal grains, fruit stones and leaves.
A part of my research focuses on how diets changed from the Late Iron Age to the Roman periods in Britain. Did the prehistoric inhabitants adopt new fruits and flavourings before they became part of the Roman empire in AD43? Were new foods restricted to the military and urban elite, or did rural dwellers also liven up their diets?
There is a large body of archaeobotanical data from Roman Britain, from 100,000s of litres of soil carefully sieved from archaeological sites over the last century. A recent research paper by Marijke Van der Veen et al. has pulled together all of the available data on plant foods from archaeological digs to show that around 50 new foods were introduced, some of which were very popular and consumed in rural sites (celery, coriander and dill), some were very rare (date, olive), and some continued to be cultivated after Roman rule had ended in AD410 (apple, pear, plum).
Yet this fascinating evidence does not seem to be reaching anyone who isn’t an archaeologist. A recent paper by Robert Witcher has shown that ideas of which plants (and animals) were Roman Introductions is well established in Britain, drawing on plant-lore, etymology, classical authors and ecology. Todays episode of BBC Radio4s Gardeners’ Question Time highlighted the different ideas about what is a Roman Introduction in horticulture and archaeobotany. In a discussion taking place at Chedworth Roman villa, several plants were identified as having been cultivated by the Romans in Britain (thyme, mint, basil, bay, hyssop, plum, apple, pear, damson). The Romans were also said to have bought new cereals with them, continued to have cultivated the old cereals (spelt and rye) and to have farmed with the mouldboard plough and double-handed scythe. Meanwhile, roman soldiers introduced Roman nettle, which they beat themselves with to keep warm.
Overall, around 30% of this is ‘correct’ based on archaeobotanical data. Yet as Witcher argues, the different types of knowledge on Roman introductions should be used to debate modern concerns over landscape, identity and ecology (alien species). As an archaeobotanist, I shouldn’t be telling people what is and what isn’t correct about aspects of our botanical heritage, but taking part in wider debates using different types of evidence.
The important point is that archaeobotanists need to get better at talking to gardeners, botanists, and everyone else about what we do know about the use of plants in the past, because it’s pretty interesting. The academic papers i’ve mentioned here are all published in journals where online access is restricted to universities, or available for around £40 a piece. The conference papers given were probably given at academic conferences. Individual archaeobotanical reports are published within books that are only available in university libraries our in journals barred by pay-walls. The archaeobotanical computer database is open-access, published in the Internet Archaeology journal, but is out of date, and not that clear to the non-specialist with plants listed by Latin name.
As a small way to approach this problem, I am going to write a series of blog posts about the archaeological evidence for ‘Roman Introductions’ to Britain. First will be the Roman Nettle – Urtica pulilifera – widely claimed as a Roman introduction, making it onto the wikipedia page for Roman Britain, but what’s the evidence?
Preston, C., Pearman, D., & Hall, A. (2004). Archaeophytes in Britain. Botanical Journal of the Linnean Society, 145, 257–294.