Category Archives: Blog

Where were the warhorses born?

An important part of the Warhorse Project is determining where the horses were born in order to reconstruct the movement of the horses throughout their lives. Key to this is stable isotope analysis, including strontium. In this blog post, we will introduce you to how the Warhorse team samples medieval horse teeth for strontium isotope analysis.

Strontium isotopes (87Sr/86Sr) can provide the location of birth in animals, including people and horses, as well as their mobility in life. As tooth enamel is formed during childhood when animals eat or drink, it records this chemical isotope from the local bedrock. This bedrock varies throughout the UK, giving us various ranges of 87Sr/86Sr, as seen below.

Map of strontium isotopic variation across the United Kingdom from Evans et al. (2010:2, Fig. 1b).

Map of strontium isotopic variation across the United Kingdom from Evans et al. (2010:2, Fig. 1b).

We take samples from horse teeth to compare to these ranges to determine where they were born within, or even outside of, the UK. The process involves the following steps, shown here with the Elverton Street horses from the Museum of London Archaeology (MOLA).

Step 1: Choose teeth from different individuals and different archaeological contexts throughout a particular archaeological site. The goal is to get an understanding of the place of birth and life histories of as many horses as possible.

Step 2: Take photographs and measurements to record the teeth before sampling, as it is a destructive method. Each sampled tooth is assigned an ID number. This is a lower (mandibular) 3rd molar from a horse that is approximately 9-10 years old, ID ELV2.

Horse tooth from Elverton prior to sampling.

Horse tooth from Elverton prior to sampling.

Step 2: Cut strips of enamel from the horse molars. We use the 2nd and 3rd molars for this. Dremels with thin, precision drill bits are used for this process.

Warhorse team member, Tess Townend, cutting strips of horse teeth with enamel in the isotope prep lab.

Warhorse team member, Tess Townend, cutting strips of horse teeth with enamel in the isotope prep lab.

Step 3: Photograph again to make sure we know which strips belong to which tooth (and which individual and context).

Horse tooth sample ELV2 after strip is cut.

Horse tooth sample ELV2 after strip is cut.

Step 4: Label, draw, and measure the tooth and the resultant strips.

Tess recording the measurements and making the drawings of the sampled teeth and strips.

Tess recording the measurements and making the drawings of the sampled teeth and strips.

Steps 5 & 6: Glue strips of enamel to paper in a ring shape, and draw a map of which teeth are where in the ring. The map is critical to be able to match the strontium isotope ratio to the particular horse.

Tess placing and mapping the strips of horse teeth.

Tess placing and mapping the strips of horse teeth.

Step 7: Fill the ring with the strips of teeth. The teeth have to fit inside the ring in order to fit in the mass spectrometer which measures the strontium.

Sampled strips of horse teeth in plastic ring.

Sampled strips of horse teeth in plastic ring.

Step 8: Pour a two-part epoxy resin over the strips, and leave to harden overnight.

Tess pouring a two-part epoxy resin over the strips of horse teeth.

Tess pouring a two-part epoxy resin over the strips of horse teeth.

Step 9: End up with a hardened resin disc with the strips of teeth in it. We have teeth from 17 different horses in two discs!

Resin disc with sampled strips of horse teeth.

Resin disc with sampled strips of horse teeth.

Step 10: Photograph the disc again to produce a final map.

Map of sampled teeth for strontium isotope analysis.

Map of sampled teeth for strontium isotope analysis.

After the teeth are sampled and prepared, Dr Alex Pryor, archaeologist and University of Exeter isotope specialist, will be undertaking the analysis using Laser Ablation Multi-Collector Inductively Coupled Plasma Mass Spectrometry (LA-MC-ICPMS) located in the Plasma Mass Spectrometer Laboratory, National Oceanography Centre, University of Southampton, UK. Once he has finished this analysis, Dr Pryor will help the Warhorse team interpret the results to gain an understanding of where the horses in the project were born and may have traveled during their lives.


Evans, J. A., Montgomery, J., Wildman, G. & Boulton, N. 2010. Spatial variations in biosphere 87Sr/86Sr in Britain. Journal of Geological Society, London 167:1–4.



Thoughts on the Chagford drift sale

By Camille Mai Lan Vo Van Qui

Each year in October, the free-roaming ponies on Dartmoor are rounded up by the farmers to whom they belong, and the foals born that year are sold at auction in the village of Chagford. This event is interesting from the standpoint of equestrian history. Those ponies are called ‘Dartmoor Hill’, meaning that they are born on the moor, and must not be confused with the Dartmoor breed. Dartmoor Hill ponies are a mix of many breeds of horses, introduced to Dartmoor over the centuries, including Shetlands used for mining in the 19th century.

Spotted Dartmoor Hill ponies at the Chagford drift sale (Photo: C. Vo Van Qui)

Keeping horses in semi-managed conditions is a traditional way of equine husbandry. In the Middle Ages, horses were already kept on Dartmoor in that manner.[1] References to this way of managing equids are found in Jordanus Rufus’s De medicina equorum (c. 1250), one of the foundational texts for equine medicine and husbandry in medieval Western Europe: he advises that foals should be born in the mountains and left to roam freely for the first two or three years of their life because, according to him, grass and freedom are what is best for their health.

De medicina equorum by Jordanus Ruffus (Photo: Wellcome Trust)

The colts are then rounded up to be taken back to the studs. There, they are tamed and broken in. Rufus gives some advice on how to capture the colts (describing, for instance, the material of the rope used to lasso them), but nothing is said about the handling of the horses or the number of people involved. This silence is no doubt due to the fact that the rounding-up and sorting out of the horses was so common and so well-known at the time that there was no point in mentioning it. Rufus was addressing himself to knowledgeable horsemen, so he only wrote down techniques and advice that were unusual at the time or the results of his own experimentations. The handling of wild[2] animals was not part of that.

A traditional event with possible medieval roots, such as the Chagford Drift sale, can provide interesting insights into the practicalities of rounding up and sorting out the horses. As a historian, it is easy to focus exclusively on textual sources. Yet, in the case of human/animal relationships, many aspects are experienced and lived and are not (or cannot) be put in writing. Therefore, when attempting to understand the relationship between humans and horses in history, contact with and observation of the animals can fill in some of the gaps in the written sources.

Chagford drift sale (Photo: C. Vo Van Qui)


During the sale, the herding of the horses was done through a succession of corrals, with minimal physical contact between the humans and animals. The horses’ instincts as herd animals appeared to be used as they were moved as groups. When humans intervened, body language was important. When the horses were touched, it was on specific areas such as the rump. Short, loud vocal cues were used to encourage the horses. Some of the handlers were brusquer, handing out small punches on the rumps of reluctant horses, while others hardly touched the animals.

Herding the young ponies between corals (Photo: H. Benkert)

Interestingly, the handling of the horse appeared to be mostly male dominated, perhaps due to the very traditional aspect of the sale, and even though women were in charge of the organisational aspect. Children, mostly boys, also participated in the handling, hinting at the generational transmission of those techniques, probably through a process of observation and mimesis. It echoes what is hinted at in Rufus’s method: the parts on which he is silent imply that they were already transmitted, either orally or non-verbally through the process of observation. As such, they did not need to be written down. Even today, no instruction manual on how to handle wild ponies exist; it is knowledge that is actively used and passed on to the next generation and it will disappear once this chain is broken.

[1] See, for instance, Charles Gladitz, Horse Breeding in the Medieval World (Dublin: Four Courts Press, 1997).

[2] Wild here refers to the horses’ experience with human interaction. Despite being mostly left to their own devices, the ponies on Dartmoor are domesticated animals and not, in fact, wild.

Pony Power!

I am Kate Kanne, a new postdoctoral research fellow for the Warhorse Project. Outside of Warhorse, my research is on early equestrianism in the European Bronze Age, and on the long-term evolution of human-horse relationships. I am thrilled to be joining this exceptional interdisciplinary team, and learning cutting-edge methods in horse archaeology to tell the story of people and horses through time. One of my first outings as a team member, was with Warhorse PhDs, Helene Benkert and Camille Mai Lan Vo Van Qui to the Dartmoor pony auction in Chagford. We were curious how these ponies were managed on these local landscapes, and how their long presence related to medieval warhorses.

Horses, and in this case proper ponies, carry in their DNA and life histories fascinating stories of human and horse history. I had never guessed when considering the Dartmoor ponies, that their story would contain everything from ancient and medieval horses, the birth of the modern polo pony, contemporary human-horse relationships, and efforts at maintaining biodiversity for the future.

A family vacation to Devon in my adolescence gave me my first wondrous experience of horses in this incredible landscape, riding at full gallop on the moors with great glee. From these memories, when going to this year’s ‘drift’ Dartmoor pony auction at Chagford, I had expected to see a sea of dark bay/brown ponies of relatively similar stature. I couldn’t have been more wrong! I confused them with Exmoor ponies who are bay, brown on dun, with black points. Welcomed by a cacophony of coat colours, and a wide variety of sizes from wee to stout, the Dartmoor ponies reflect their long history, as well as the human proclivity for valuing eye-catching coat colours, those of which also bring a higher auction premium (Figures 1, 2, & 3).

Colourful Ponies at the Chagford Auction

Colourful Ponies at the Chagford Auction

Colourful Ponies at the Chagford Auction

Figures 1. 2. & 3. Colourful Dartmoor Ponies at the Chagford Pony Auction (Figure 1 & 2 photos by Helene Benkert).

Figure 4. Wee Dartmoor Pony at the Chagford Pony Auction.

Black Dartmoor Hill Pony Stallion with Moustache!

Figure 5. Black Dartmoor Pony Stallion with Moustache!

I was also surprised to learn that the herds of free-roaming ponies are all owned, with the owners (Registered Commoners) allowed to graze their herds on the Commons (Figure 6).

Dartmoor Ponies on the Moor

Figure 6. Ponies grazing on Dartmoor Commons (Wikimedia Commons, Andrew Rendell CC-BY-2.0))

For those less familiar with British law, Common Land is privately owned with ‘Rights of Common’ for land used most often to graze livestock. Enshrined in the Charter of the Forest in 1217, in pushback against rights of access to the royal forest for free men eroded by William the Conqueror and his heirs, the Commons were designed to sustain poor rural farmers who did not own their own land with access to pasture, wood, and other resources. Now, over one-third of England’s moorland is common land, 3% of land in England, and includes famous landscapes in Dartmoor National Park, the Lake District, Yorkshire Dales and Shropshire Hills (Dartmoor Commoners’ Council) (Figure 7).

Common Land on Dartmoor

Figure 7. Common land on Dartmoor National Park (Dartmoor Commoners’ Council).

From the Dartmoor Hill Pony Owner’s Club, we learn that, “A Dartmoor Hill Pony is one bred on the Commons of Dartmoor by a Registered Commoner, whose sire and dam run on the said commons. This ensures that the sire has been inspected and approved by the Dartmoor Commoners Council as a suitable stallion to run on the Commons”. Current estimates suggest there are around 1100 mares producing roughly 900 foals per year. The ponies are rounded up from the multiple Commons in Dartmoor annually in October. They are then counted, checked, with some sold at the traditional pony Drift Sale in Chagford, Devon, held by the Dartmoor Hill Pony Association, with volunteers of Friends of the Dartmoor Hill Pony charity, overseen by Moorgate Vets, and facilitated by Rendells Auctioneers (Fig. 8).

Chagford Drift Pony Auction

Figure 8. Chagford Drift Pony Auction.

The drift auction originally was a method by which illegally grazing or unowned animals would be auctioned in the Forest of Dartmoor. Now it serves to keep the ponies at a manageable number for their owners and the moors, and provides farmers with an additional form of income. The value of the Dartmoor ponies lies in their use as excellent, sure-footed, and hardy riding or driving ponies, as well as a keystone conservation management species (Figure 9).

Dartmoor Ponies Conservation Grazing

Figure 9. Dartmoor ponies conservation grazing at The Garrison, St. Mary’s. They were introduced to to restore Scilly’s maritime heathland (Bob Embleton / Conservation Grazing The Garrison, St. Mary’s / CC BY-SA 2.0)

The Dartmoor ponies represent the decedents of the earliest domesticated horses in England, attested to from the Bronze Age (Bendrey 2012) (Fig. 10).

 Dartmoor Pony at Bronze Age Archaeological Site

Figure 10. Ponies on Dartmoor at a Bronze Age archaeological site (© Charlotte Faulkner in Packham 2020).

At Shaugh Prior in South Dartmoor, pony hoof prints were found in association with cattle and sheep hoof prints, dated to the Bronze Age, approximately 1500-1350 BC (Fig. 11).

Pony, cattle, and sheep hoof prints under excavation on Shaugh Prior, South Dartmoor

Figure 11. Pony, sheep, and cattle hoofprints under excavation on Shaugh Prior (© Ossiie Palmer in Packham 2020).

Many different types of horses were brought to the British Isles in subsequent periods, breeding with the initial stock that became supremely adapted to life on the moors. Horses became increasingly important for battle in the Iron Age and Roman period as mounts, and famously as chariot horses. The quality of these smaller ponies was noticed by the Romans, who took them back to Italy to improve their native stock (Trew 1953). Ponies are well known archaeologically throughout the Early Middle Ages, into the Late Anglo-Saxon / Early Tudor period of the Warhorse project and beyond. Withers height indicates that horses from the Saxon and Norman periods (5th–12th centuries) were ponies by modern standards, less than 14.2 hh or 1.48 m, generally averaging 12.3-13.2 hh (Ameen et al. 2021).

The first historical record of herds of free-roaming ponies in England is from the will of the Merican thane Wulfric, who in 1002 decreed “’…I leave to the monastery at Burton one hundred wild horses and sixteen broken stallions…” (Dent and Goodall 1968:53) (Figure 12).

Figure 12. An 18th pencil drawing of the stained glass window at Hall Hill, Abbots Bromley, Staffordshire, England, showing the Anglo-Saxon nobleman Wulfric Spot (died c.1004) (Wikimedia Commons, Public Domain).

Shortly thereafter, the Saxon “Ælfwold, Bishop of Crediton, in 1012, bequeathed ‘to every man of my household his mount which I have leant him… Aelfwold also left a stud of ‘wild horses’ running on the Buckland estates near Dartmoor, besides ten individual bequests of saddle-horses, and to his lord the heriot of four horses with armour…” (ibid). In the Domesday Book from 1086, held at the Exeter cathedral, in the owned property survey of the UK conducted for William the Conqueror to calculate tax, the wild ponies are discussed, but there was some uncertainty if they should be taxed at all due to the fact that they were semi-wild, but owned herds and the right to graze them on Dartmoor was already in place. Clearly there would have been an awareness of the herds by the those producing horses for medieval cavalries, as well as the rulers themselves.

Henry VII directed the improvement of native stock for his warhorse retinue in 1535, requiring by law that mares should be bred with stallions that were 14 hands or greater, or the owner be fined 40 shillings. His disparaging remarks deemed the ponies of Dartmoor ‘little horses and nags of a small stature’ and a ‘vile and paltry breed’.” However, given the remoteness of Dartmoor, Henry’s policies were likely not enforced. By the 16th century, ponies of all colours and sizes were recorded.

The ponies were used extensively use as pack horses, and in the tin mines, as well as an “all-rounder” for riding and driving. In 1855, polo came to Britain from India significantly affecting pony breeding. By the 1860s, under the influence of the Prince of Wales, systematic efforts were undertaken to ‘improve’ ponies from the Dartmoor herds with outcrossing with Thoroughbreds Arabians, Hackneys, Welsh, and Fell horses to produce an ideal sport pony. Dartmoor ponies were foundational to the formation of the Polo Pony Stud Book in 1899 (Fig. 13). A famous polo pony named “The Leat” was purchased by Miss Calmady Hamlyn in 1925, the by an Arabian stallion, out of a Dartmoor pony x Hackney mare (Friends of the Dartmoor Hill Pony). “The Leat” was the first pedigree Dartmoor Pony (Hendricks 2007). In 1899, The Dartmoor Stud Book was started as a section of the Polo Pony Stud Book, becoming a separate entity in 1925 for registered Dartmoor Ponies.

Polo Pony Stud Book, made possible by Dartmoor Hill Ponies

Figure 13. The Polo Pony Stud Book, made possible by the Dartmoor Ponies.

Dartmoor ponies continued to live semi-wild on the moor at the same time, refusing efforts to conform with the fashionable breeding of the early 20th century. They remain genetically distinct from their registered counterparts, with key signatures that can be used to distinguish them, as well as signs of genetic selection potentially linked to environmental adaptation (Hegarty et al. 2017). Now preserved for their heritage value, as well as their key contribution to biodiversity, the Dartmoor ponies continue to provide enjoyment on and off the moors, through their use in riding and driving, as companions, or as conservation grazers.

The story of the Dartmoor ponies is the story of horses in England. In them we see the relationships of horses and people to the land and place, the negotiations of commoners with, and resistance to, people in power, the centuries long traditions of families working together in local practices of breeding and management, and the recognition of the vital place of the Dartmoor ponies in the past, present, and future of Dartmoor and its people.


Ameen, C., Benkert, H., Fraser, T., Gordon, R., Holmes, M., Johnson, W., Lauritsen, M., Maltby, M., Rapp, K., Townend, T., Baker, G. P., Jones, L. M., Vo Van Qui, C., Webley, R., Liddiard, R., Sykes, N., Creighton, O. H., Thomas, R., & Outram, A. K. 2021. In search of the ‘great horse’: A zooarchaeological assessment of horses from England (AD 300–1650). International Journal of Osteoarchaeology, 111.

Bendrey, Robin. 2012. From wild horses to domestic horses: a European perspective. World Archaeology 44(1):135-157.

Dent, A.A., and Daphne Machin Goodall. 1968. A History of British Native Ponies: From the Bronze Age to the Present Day. London: J.A. Allen.

Green, Peter. 2016. The free-living ponies within the Exmoor National Park: their status, welfare and future. A Report to the Exmoor Moorland Landscape Partnership.

Hegarty, Matt, Nicola McElhinney, Emily Ham, Charly Morgan, Clare Winton and Rob McMahon. 2017. An evaluation of the genetic relationships between the Hill Dartmoor and the registered Dartmoor Pony Breed. Report prepared for the Friends of the Dartmoor Hill Pony. Aberystwyth University.

Hendricks, Bonnie. L. 2007. International Encyclopedia of Horse Breeds. Norman, OK: University of Oklahoma Press.

Packham, Chris. 2020. A Keystone Species: The Dartmoor Hill Pony. Springwatch. Photos from: Bronze Age Hoofprints © Ossiie Palmer; Ponies at Bronze Age settlement © Charlotte Faulkner.

Trew, Cecil G. 1953. The horse through the ages. London: Methuen.

Warhorse and Public Engagement: Pop-up Shop of Science and Culture

An empty shop opposite Princesshay in Exeter city centre might seem an unlikely place to see the AHRC Warhorse team sharing their enthusiasm for the project, but a Saturday in late September witnessed this location transformed into a ‘Pop-Up Curiosity Shop of Science and Culture’  by Agile Rabbit in which members of the public could engage for free with demonstrations, talks and activities. Styled the ‘Maketank Gallery’, the space was transformed into a veritable research bazaar on Thursday afternoon, with stalls and displays showcasing cutting-edge work, ready for Saturday’s main event.

Members of the Warhorse team with the stall prepared and visitors eagerly awaited…

Public outreach is integral to our project, and the Covid crisis has presented some major obstacles with our planned activities over the last 18 months, although we did take advantage of digital methods of delivery to organise training workshops that were covered in an earlier blog. The Pop-Up Shop provided us with a fabulous opportunity to showcase our project, its team and some of our findings and activities to members of a very diverse public. We produced a table-top exhibition from which we delivered ‘show and tell’ activities throughout the day to a great range of visitors, from shoppers opportunistically popping in, to some formidable specialists on all things horsey…

Our stall presented an array of materials covering different aspects of our work — particularly horse bones and equine material culture. The public were particularly drawn to — and sometimes mystified by — the horse bits, and really enjoyed handling our collection of horse bones, including a skull. A fragment of fused horse spine — the condition known as ‘bamboo spine’ — from the Department of Archaeology’s collections  drew some gasps as members of the public realised the damage that carrying heavy loads could wreak on an equine body.

A fused horses spine from an archaeological context. Many people commented on the pain the horse must have felt: “really gruesome but really interesting” in the words of one young visitor.

A ‘design your own horse harness pendant’ kit proved an instant hit with Exeter’s younger folk, with many producing their own heraldic designs and hearing strange words such as ‘tincture’ and ‘argent’ for the first time!

The ever-popular ‘make your own horse pendant’ kit, aimed at children.










One of the most engaging questions we posed members of the public was to guesstimate how tall a medieval warhorse would have been. With the help of a special horse measuring tape, marked out in hands, we were able to draw on the results presented in our soon-to-be-published paper in The International Journal of Osteoarchaeology to show how medieval horses were far smaller than the Shire horse-sized beasts of popular imagination. “I had no idea a medieval warhorse could be so small; it’s fascinating to hear that historical films can get it so badly wrong!” proclaimed one enlightened visitor. Another commented on how interesting it was to hear about archaeologists “hunting down the mythical beast” that is the medieval warhorse.

One of the most exciting aspects of our project’s development is the way that the research team has grown to encompass students at different stages of their journeys, and it was excellent to see current University of Exeter students who had carried out research on warhorses as part of their studies join in alongside our doctoral students, post-doctoral research staff and investigators.

Overall, the event attracted well in excess of 300 members of the public and proved a fabulous showcase for our work. We sincerely hope that we are able to seize on similar opportunities in the future.

Members of the warhorse project team talking with some engaged members of the public at the pop-up exhibition stand.

Equine objects – the horse in artefacts (Part 2)

This is the second part of a post in which we have been considering the horse and rider in artefacts in order to develop our sense of the horse’s contemporary significance. In Part 1 Laura Jones examined a particular object type – a padlock in the form of a horse – for which our knowledge mostly derives from metal-detected finds recorded by the Portable Antiquities Scheme (PAS). In this part Rob Webley continues the analysis of PAS records, considering a copper-alloy sheath chape which features a highly intriguing design.


Chapes are metal components which protected the terminal of a scabbard or a sheath, and, ultimately, the owner from accidental damage from their blade. Chapes made of folded sheet metal, and possessing one attachment arm (such as the one shown below), are usually considered to have come from leather sheaths for knives. Beyond their function, a chape could help decorate a sheath, with openwork details interplaying with the colour of the leather beneath for aesthetic effect. Here the focus is on one of the most highly decorated of a group of folded knife sheath chapes, as on one face a horse and rider are depicted (see image below, to right).

Medieval sheath chape (PAS: SUSS-03FD90) (Image courtesy of the Portable Antiquities Scheme)

Medieval sheath chape (PAS: SUSS-03FD90) (Image courtesy of the Portable Antiquities Scheme)

Although these chapes from knife sheaths, being folded, featured two faces which could be decorated, only this single type seems to have borne two different designs. Recently classified by museum officer and researcher Ben Bishop (2020) within his Class D, we will concern ourselves only with the face that shows a knight on a horse, the other face having been subject to various interpretations. In the following description of the knightly scene my own interpretation is offered: even on the clearest example of such a chape, found at Angel Court in the City of London, the features may be construed in different ways. This interpretation is shown in the coloured image below: the knight has a pointed helmet, carries a kite-shaped shield on their left arm and wears protective clothing – possibly a gambeson (a padded jacket) – which is visible below the shield. In their right hand they hold a battle axe, the shaft of which rests on their shoulder. The horse’s forelegs rest on the ground such that the rider’s feet virtually touch the ground. I have emphasised the horse’s harness, with the reins held by the rider in their left hand and with both a breast- and rear-band attached to the saddle to give it stability. This is a different interpretation to those who have suggested that marks on the horse’s flanks might represent a caparison (protective cloth covering) – a point to which we will return.

Interpretation of the design of a medieval sheath chape (PAS: SUSS-03FD90) (Image courtesy of the Portable Antiquities Scheme)

Interpretation of the design of a medieval sheath chape (PAS: SUSS-03FD90) (Image courtesy of the Portable Antiquities Scheme)

Regardless of the precise details shown on these chapes, the fundamentals present a puzzle regarding its dating and the intriguing knightly depiction. Dating this type of chape is made more difficult by the fact that none of the over 40 examples known has been found in a stratified context. A small clue is offered by the terminal of the Angel Court chape, the only surviving arm terminal for this group. It is sub-oval with a bifid end (i.e. split in two parts) and can be paralleled by terminals present on examples of Bishop Class A chapes which can be tentatively dated to the first half of the 12th century. Such dating may be compared with the internal evidence of the chape itself which depicts attire typical of an 11th- to 12th-century knight, as seen, for example, on the Bayeux Tapestry (c. 1070s) or the Winchester Bible (c. 1160/1170s). Of the elements depicted, it is arguably the kite-shaped shield that is the most diagnostic of date, with its rounded top suggesting a date before the mid-12th century. Taking the strands of evidence together, I would favour a date in the early 12th century for these chapes, though various of the elements depicted admittedly do not provide such precise dating, being plausibly slightly earlier or later.

At first glance, an early 12th-century date does not sit perfectly with two particular elements described: the possible caparison and the battle axe. Iconographic evidence places the advent of the caparison (or cloth trapper) in the 1140s, based on its depiction on the seal of Simon de Senlis, Count of Huntingdon and Northampton. However, their conspicuous absence from seals until the end of the 12th century, if not later, suggests two options: that caparisons were more common earlier than we thought, or that something else is shown by these engraved lines – patterns on the horse’s hair (elsewhere, punched dots might show dappling). The battle axe might seem a typically 11th-century weapon were it not for mention of one in Henry of Huntingdon’s description of the Battle of Lincoln (1141), right at the end of the first half of the 12th century. The axe is arguably the most striking part of the ensemble, and makes this an atypical knightly depiction. The medieval military historian David Nicolle (2011, 39) has speculated that these chapes might have belonged to returning English Varangian guards, but their quantity might mitigate against this suggestion (see also map, below). Overall, the arguments constructed regarding the dating and depictions on these chapes represent but one interpretation, naturally open to challenge – if everything was straightforward it would not be so fun!

Map of Bishop Class D sheath chapes (by Rob Webley)

Map of Bishop Class D sheath chapes (by Rob Webley)

The distribution of these chapes may be contrasted with the padlocks discussed in Part 1 as the chapes are a strongly southern phenomenon, and notably absent in the north of the country. In part, this may be put down to the use of knives with sheaths furnished with chapes which, based on current recording of surviving examples, seems to be a more southerly feature overall. On the other hand, in the 12th century this depiction of elite equestrianism may have been more culturally legible in the south of the country. Taken together, the objects discussed in this post offer new insights into perceptions of medieval horses and riders, though how ‘popular’ (as opposed to elite) they were would benefit from further work, as well as more examples found in context!

Bibliography (Part 2)

Bishop, B. 2020. Early-medieval and medieval bifacial sheath fittings, Sleaford: Finds Research Group Datasheet 53.

Nicolle, D. 2011. The Fourth Crusade 1202-04: the betrayal of Byzantium, Oxford: Osprey Publishing.

Equine objects – the horse in artefacts (Part 1)

One means of examining the ways in which the horse pervaded the medieval mind is to consider how often and in what ways it was represented in material culture. It is well known that the horse featured widely in medieval manuscript art, on wall hangings and in architectural sculpture. Its presence as/on/in artefacts (in two or three dimensions) has perhaps been less fully discussed, with the exception of the equestrian seal and the aquamanile (hand-washing vessel) in metal; both objects show knights on horseback and both hold elite associations. As a potential way into examining more popular reception of the medieval horse, in this blog post we will consider a relatively untapped source for exploring horse art – the corpus of metal-detected finds recorded by the Portable Antiquities Scheme (PAS).

In a previous post on the PAS dataset we introduced the range of actual horse equipment – horseshoes, harness pendants, stirrup-strap mounts, and so on. But within the PAS database, now numbering over 1.5 million artefacts, there are also a number of objects which either show, or take the form of, a horse, but which are probably not items of equestrian equipment. Across the two parts of this post, Laura Jones and Rob Webley examine two groups of medieval object that loom large within the set of ‘equine’ (rather than ‘equestrian’) objects – the subject of Part 1 is a horse-shaped padlock moulded in the round. In Part 2 the focus turns to a chape (i.e. the metal point of a scabbard or sheath) with relief decoration that includes a horse and rider.


Zoomorphic padlocks reported to the PAS database commonly represent horses, with just under 40 examples of such padlocks recorded to date. They are elongated, in a moulded case, with holes for the padlock bolt in the rear end of the horse and the keyhole situated at the horse’s chest; the padlock bolt fits into a hole located in the horse’s head. Very rarely, examples are preserved with the bolt in situ (see SUR-D02FD3 below).

Medieval zoomorphic padlock (PAS: SUR-D02FD3) (Image courtesy of the Portable Antiquities Scheme)

Medieval zoomorphic padlock (PAS: SUR-D02FD3) (Image courtesy of the Portable Antiquities Scheme)

Regarding decorative details, these padlocks have lines that depict the horse’s saddle; others depict bridles, bands and even the mane of the horse – some decorative examples can be seen below.

Highly decorated medieval zoomorphic padlock (PAS: SOM-ABF421) (Image courtesy of the Portable Antiquities Scheme)

Highly decorated medieval zoomorphic padlock (PAS: SOM-ABF421) (Image courtesy of the Portable Antiquities Scheme)

Medieval zoomorphic padlock depicting a harness pendant (PAS: WAW-565B1A) (Image courtesy of the Portable Antiquities Scheme)

Medieval zoomorphic padlock depicting a harness pendant (PAS: WAW-565B1A) (Image courtesy of the Portable Antiquities Scheme)

Within these horse-shaped padlocks, there are certain distinct patterns – enough to suggest potential groupings of types, though no typology has been created as of yet. Many padlocks are rectilinear in profile such as SUR-D02FD3 (above). A particular challenge is how to group types of decorative zoomorphic padlocks such as SOM-ABF421WAW-565B1A and LEIC-CE40BE because of the differences in the stature of the horse and equestrian equipment depicted on these padlocks in comparison to the rectilinear padlocks. Lastly, an ever-present question is how do we classify a given padlock when we have to take into consideration damage to the objects or unique features?

Medieval zoomorphic padlock depicting girth straps (PAS: LEIC-CE40BE) (Image courtesy of the Portable Antiquities Scheme)

Medieval zoomorphic padlock depicting girth straps (PAS: LEIC-CE40BE) (Image courtesy of the Portable Antiquities Scheme)

A good place to start is to focus on the equestrian equipment depicted on the padlocks such as the bridle on SOM-ABF421 (above) that is demarcated by two crossed grooves on the horse’s head. If we contrast this to the horse’s head on LEIC-CE40BE (above), which shows compression of the object (possibly, as the record suggests, by the padlock’s bolt), can we be certain that there were no other decorative or unique details? Also up for discussion is the shape of the horse. Some padlocks are described on the PAS as stocky, which could allude to a pack horse, and others look almost taller in stature and tend to be what I like to term ‘dynamic’ (legs in pose or active stances). However, it is important to note that it is common for either/both front and hind legs to be broken. This changes our perception of the horse’s stance and stature and makes it hard to decide on a specific classification. In addition, there are unique details depicted on these padlocks such as the harness pendant on WAW-565B1A (shown on the band on the upper sides above the front legs) and the girth straps below the saddle on LEIC-CE40BE – commonly seen in iconography of the ‘warhorse’. For now, we shall leave the grouping of types to future discussion and refer to the general classification given to them in a wider discussion of padlocks by Lewis (2016, 168-9) – Type V ‘zoomorphic’ padlocks.

Regarding the date of these Type V padlocks, we can use excavated examples to ascertain a rough date range. A padlock was found within Winchester’s medieval city within a context that included a Clark type 2B horseshoe (Goodall 1990, 1011), the horseshoe’s typological dating of c. 1150-1225 supporting the dating attributed to the site phase (12th to early 13th century). On the PAS, the relevant Finds Recording Guide provides a broad date range for these padlocks of between c. 1100-1400 because of similar models from the Middle East that date between the 13th and 14th centuries. The evidence from Winchester, plus the way in which these padlocks are decorated, makes a 12th- to 13th-century date for them plausible.

Lewis (2016) refers to the ambiguous nature of these zoomorphic padlocks – we do not know their specific function nor do we know their connection to medieval society. One possibility is that they are connected to high-status sites within the English landscape – locations such as manor houses, castles and palaces. Jervis (2011) was influenced by zoomorphic pottery found at high-status sites in Southampton and applied the idea of zoomorphic symbolism connected to wealth to these padlocks. This assumption is linked to elite symbolism of animals related to the hunt – by displaying them through objects, it may reinforce a message of wealth, hence reinforcing their connection to elite sites. Of note is the distribution of these padlocks across England; as you can see from the distribution map, most of the padlocks are situated in rural remote locations (apart from a few clusters around urban sites such as Winchester, York and Norwich).

Map of Lewis Type V padlocks (by Laura Jones)

Map of Lewis Type V padlocks (by Laura Jones)

When researching the context of the find location spots for these padlocks, all have medieval manorial links (medieval manor houses proximal/directly on or to the location of the padlocks). Interestingly, some find spots such as the above records from Warwickshire and Surrey have close connections to horses; the Warwickshire record was found in a medieval settlement tied closely to the area known as the Vale of the Red Horse (an area between Edgehill and the Northern Cotswolds); the Surrey record has surrounding horse-related place names such as The Paddock and elite place names like Kingwood Common and Earl’s Wood (all of the above forming the 12th to 13th-century parish where the de Grey family manor is situated). The exact function of these padlocks (beyond fastening a casket) is hard to ascertain, regardless of what the distribution map or the context of location spots may reveal to us. The links could support the theory of these objects being associated with the elite however the contextual information of a few records are not definitive evidence that they are solely linked with elite sites (and the location of the Winchester example is certainly not in a high status one). More research would have to be conducted into these sites by looking at further Type V padlock records, alongside additional historical records, to reveal more about the connection between medieval society and these padlocks, which may even reveal more about their precise function.

To be continued

Bibliography (Part 1)

Goodall, I. H., 1990. ‘Locks and Keys’, in M. Biddle (ed.), Object and Economy in Medieval Winchester: Artefacts from Medieval Winchester, Oxford: Winchester Studies 7, 1001-1036.

Jervis, B., 2011. ‘Placing Pottery: An Actor-led Approach to the Use and Perception of Medieval Pottery in Southampton and its Region c. AD 700-1400’, unpublished PhD thesis, University of Southampton.

Lewis, M., 2016. ‘Mounts for Furnishings, Padlocks, and Candleholders: Understanding the Urbanization of Medieval England through Metal Small Finds Recorded by the Portable Antiquities Scheme’, in B. Jervis, L. G. Broderick and I. G. Sologestoa (eds), Objects, Environment, and Everyday Life in Medieval Europe, Turnhout: Brepols Publishers, 159-185.

The Horses of the Royal Household under the Three Edwards

In this latest post, Dr Gary Baker explains the equine organisation within the royal household.

Much of what has been written on king’s horses in medieval England has focussed upon the royal studs: the locations in which the monarch’s warhorses of the future were bred. What is often neglected is the other part of the royal equine network: the horses within the royal household. The relative lack of scholarship on the royal household horses is important, because it was the royal household into which the horses bred in the king’s studs were sent, used either as mounts for the king and his inner-most associates or else requisitioned for use in the household’s myriad of departments. Also, whilst the studs provided warhorses for the king for the most part, some of these animals deemed unfit for war were used by the various household departments as pack horses, messengers’ horses, or everyday riding horses.

The royal household on the move. Horses were used for a variety of purposes in the household such as riding, baggage, for messengers, and pulling important persons, as here

The organisation of the royal household in the early fourteenth century was laid out in a household ordinance of 1318. The ordinance specified five great offices of the household: the steward, the keeper of the wardrobe, the chamberlain, the controller, and the cofferer. The king’s horses appear within the Marshalsea, a ‘sub-department’ under the jurisdiction of the keeper of the wardrobe, who acted as the household’s treasurer. The Marshalsea (not to be confused with the Marshalsea Court which administered justice in the household and twelve miles around the itinerant household as it moved around the country) was staffed by several senior officials, each with specified functions related to the maintenance and care of the king’s horses. There was the chief clerk, whose job was to account for the spending of the chief purveyor of the avenary, the latter of which was charged with obtaining foodstuffs and other goods for the horses. The day-to-day care of the horses was supervised by two sergeants known as harbingers, each with a deputy, and three sergeant-marshals.

The medieval household was a vibrant place, with many courtiers and officials present at any one time

The 1318 ordinance does not provide the names of any of these equine officials, but we can identify a good number of them by looking at near-contemporary wardrobe accounts. The wardrobe book of Roger de Waltham, Keeper of the king’s wardrobe for 1 May 1322 – 19 October 1323, is a perfect example in this regard. Members of the royal household were provided with an annual allowances for clothes (robes) and shoes, with senior members of the household entitled to allowances twice-annually in winter and summer. Waltham’s 1322–23 account, which straddles three of Edward II’s regnal years (15–17), therefore covers the payment of garment allowances in the summer of 1322, and the winter and summer of 1323, and lists by name members of the royal household in receipt of these allowances. From these lists we can name virtually all the members of the royal equine establishment at this time.

Senior officials of the Marshalsea in receipt of allowances for robes, 1322-23

In the summer of 1323, for example, the wardrobe book names 32 carters in receipt of an allowance to buy shoes. These men worked throughout the household’s various departments along with 85 valets, sumptermen, palfreymen, and keepers of the horses of the king in the household’s stables . These 85 men are listed within the departments in which they worked, such as one man managing the pantry’s horse and another the banquet hall’s horse; beasts most likely used to carry the various equipment of those departments. The allowances for robes are even more revealing, as they also name several senior equine officials. The chief clerk of the Marshalsea at the time was Simon Eycote; Maurice Drageheved was the sergeant of the palfreys; Master John Gilmyn the marshal of the horses; John Mildenhale was senior farrier; Giles de Toulouse, was keeper of ‘certain horses of the king within the household’; Hugh de Beaurepaire, was keeper of certain of the king’s horses outside the court (extra curiam); and John de Reading, was sergeant and harbinger of the sumpters and carthorses. In addition to these senior officials the lists of those in receipt of robes also name 4 purveyors, 2 harbingers, and 5 farriers, along with 30 sumptermen working in diverse other departments. Unfortunately no overall totals for the number of horses in the household is provided, but historians have argued that since each groom was usually in charge of one horse, there could well have been anywhere between 100 and 200 horses present at any one time; a combination of the king’s warhorses and also the various horses required for baggage, messengers, and transporting the household’s people and goods round the country.

Horses were bred in the royal studs and probably looked mean, like this lot!

At this stage the external network of the king’s studs was also accounted for in the wardrobe (it would later do so at the exchequer). In 1322–23 we can thus see the external stud network operating in parallel to that of the household. The aforementioned Hugh de Beaurepaire received £344 9s 7½d for the period 1 May 1322 – 7 July 1323 for his wages and the upkeep of an unspecified combination of 40 destriers, coursers, and other horses at the royal stable at Reading, and another £71 1s 10d for keeping 30 horses, again at Reading, from 8 July to 19 October. At the same time the royal southern studs were in the custody of Brother John de Redemere, a Dominican friar. Redemere received £210 1s 1d for his wages and for those of various stable hands, and the costs of keeping 4 stallions, 23 mares, and 41 foals at Woodstock; 2 stallions and 24 mares at Cornbury; 2 stallions, 19 mares, and 6 colts at Risborough; 32 colts at Odiham; and 4 stallions, 71 mares and 26 colts at Rayleigh (254 animals in total), and an unspecified number at Windsor, from 1 May 1322 – 7 July 1323. In the north, first Richard de Merksale (9 January – 20 March 1323), and then John de Neusom (20 March – 19 October), received £253 10s 5d for their expenses, wages, and the wages of John de Derlington (Darlington), Marshal of the horses, one farrier, and other valet’s and pages, keeping 28 of the king’s ‘great horses and a lesser number of other horses’ in Yorkshire.

Whilst this is only a snap-shot in time it should hopefully show that there was more to the royal equine administration than the horses in the king’s studs, and as the household grew so too did the number of horses required for its personnel and baggage. The royal household on the move, with hundreds of people and animals on the move must, therefore, have been a very impressive sight.

Summary blog of work over the last year

A year is a long time in research, and the interval between the last project summary blog in January 2020 and this update twelve months later has seen some seismic shifts in ways of working for a project such as ours due to the ongoing Covid pandemic.  As well as confronting some obvious challenges, our project has put in place mitigations and work plans to drive the work forward and make exciting progress and new discoveries on several fronts.

A meeting of Warhorse project staff 2021 style

The Zooarchaeological Work Package has developed in particularly innovative ways during the crisis in order to mitigate some of the difficulties around many museums and archive collections being closed for long periods of time, which has prevented ready access horse bones for analysis.  Carly Ameen, our Project Officer, has led the way in building a large digital dataset of existing measurements of horse bones. Containing in excess of 6,000 records sourced from a wide network of zooarchaeological collaborators, this dataset is proving a rich and remarkable platform for deep analysis of horse metrics. Preparing this resource has been time consuming and involved, including extensive cleaning of records and streamlining for consistency, but we are quietly confident that it will enable us to model changes in horse size through the centuries in a way not previously possible. Preliminary analyses are both promising and intriguing, and the database has the potential to bring out contrasts between horse remains from different types of sites (such as castles, religious communities and settlements) and between the rural and urban worlds, as well as through time. We are already preparing a manuscript for a paper to present some headline results. Despite the issues of accessing museums, the team has still managed to locate specimens of horse bones from some choice sites that have been transported to our facilities in Exeter for sampling and analysis (see this earlier blog post). The materials have ranged widely — from early medieval horse bones derived from Whitby Abbey on the coast of Yorkshire through to early modern examples from Bradgate Park, Leicestershire — and our work has involved X-Ray recording and the extraction of samples for genetic analysis that have been mailed off to our partner’s laboratory in Toulouse. We await seeing the preliminary results…

Progressing our Material Culture Work Package during the Covid-19 pandemic has been a little less challenging than some other areas of the project given that it has as its core the analysis and enhancement of a digital resource that is fully online — the Portable Antiquities Scheme Database. Aided by Laura Jones, a University of Exeter Undergraduate Archaeology student on a year-long work placement with our project, Rob Webley has been driving forward work on cleaning and enhancing digital records relating to equestrian finds and on the spatial analysis of findspots. The project has already contributed to the development of the online Finds Recording Guides created by the Portable Antiquities Scheme — this link gives a good flavour of the work and of the types of horse gear that metal detectorists around the country are finding every day. Sometimes copper-alloy components can help us analyse equipment otherwise dominated by iron — an article published recently by Rob Webley suggests that a previously mysterious type of mount was in fact used on (iron) bridle bits. Harness pendants have been a particular focus of work, and the team have broken new ground in exploring representations of heraldry on these tiny but illuminating artefacts (see this earlier blog post) and studying their patterning in the landscape. One particularly tantalising possibility that we are exploring is that some documented medieval tournament sites preserve diagnostic artefact scatters including horse harness pendants. In the last year we have also made great initial strides with measuring armour, starting with the stored collections of the Royal Armouries in Leeds (see this blog post), but this is an element of the project that will have to stay in abeyance until museums and collections open up.

The History and Landscapes Work Package, led by our team at the University of East Anglia (Rob Liddiard and Gary Baker), has been making remarkable progress despite the challenges presented by the non-accessibility of some key historical archives during the Covid pandemic. The preliminary work on photographing records of royal horse accounts held by the National Archives in Kew that was carried out pre-pandemic has meant that some initial analyses have been possible — for example see this blog post on what these documents reveal about the maintenance, care and feeding of horses in the fourteenth century. One of the aspects of the project that excites us most is the potential for interplay between different categories of evidence, and a particularly novel aspect of this work package is the question of how we can relate documentary records of horse management to the physical evidence of the landscape. Excellent progress has been made on mapping for the first time the nationwide distribution of horse studs as it evolved through the centuries, and on reconstructing the hinterlands on which individual studs drew upon for resources. Hard details on the appearance of stud infrastructure is proving frustratingly elusive in the documents but we are hopeful that future combing of select archives will produce results, while the evidence of place-names provides another clear area of potential: see this blog for some early thoughts on what place-names can tell us about medieval equine landscapes. Over the course of the next year we will be selecting sites for fieldwork, including geophysical surveys, in our hunt for stud sites, and following up other leads in the documentary record.

With a series of journal articles in different stages of development and some already out: see for example: and the outline of our project monograph developing fast, the team is keeping a keen eye on outputs and even follow-on funding opportunities to help build a legacy of our project. Project team members are also active on the international conference circuit, with highlights among planned presentations over the coming months being a paper at the EAA (European Association of Archaeologists) conference and the Battle Conference on Anglo-Norman Studies.

Do look out for future posts as the project develops in these directions and others.

Reconstructing Past Horse Conformation

I am Karina Rapp, and I am an Honorary Associate Research Fellow, Zooarchaeology Masters graduate, and PhD hopeful from the University of Exeter. I am an active horsewoman, equine journalist, and the president of a nonprofit registry that promotes a rare breed of historic cavalry horse, supports global rare breed conservation efforts, and educates the public about the benefits of genetic diversity. I greatly enjoy hands-on research that has the ability to change people’s perspectives on history, especially when it includes faunal remains. The Warhorse project ticks all the boxes for me, and I am so grateful to contribute towards a small part of it! My area of interest is equid conformation (functional shape), and I enjoy exploring ways to adapt modern conformation studies to archaeological remains and skeletal biomechanics.

Archaeological horse bones and material culture (such as bits and armour) can be compared to measurements from modern live horses to give a potential idea of the size and body type of past horses. Several basic types of horse exist, such as small ponies, average horses, and draught horses. This is relevant for reconstructing what the Medieval “Great Horse” actually looked like, and for the very interesting history of selective Medieval horse breeding. By developing a baseline of reliable data from modern horses (including important information such as their height and body type) we can see where ancient horses lie on the allometric spectrum between small ponies and large draughts.

Working remotely in the United States, I have traveled to four barns in southern Idaho to study a variety of animals: working ranch horses, children’s Welsh ponies, show jumping warmbloods, and Belgian draughts used under harness for agricultural work. Measuring a living breathing creature can be more complicated than old bones, though! Hard-edged callipers and bone boards are traded out for flexible seamstress measuring tapes and custom devices for getting around a horse’s muscular curves. Quiet measurements of bones deep in a well-lit lab are traded for dusty barn aisles, quick-moving targets, and sometimes unwilling 1000 lb. subjects in small spaces. The bones of living animals are encased in cartilage and muscle and tendons; it is important to understand that living measurements will never match up exactly with the dry measurements of skeletal materials. Measurements are therefore less about exact bony markers and more about best-guess estimates of chosen data points that correlate as close as possible to certain bone lengths.

Basic body measurements of the horse (such as face and leg measurements) are obtained with a flexible seamstress tape, and can help relate horse size to custom-fit equine armour. Bit sizers are used in the modern horse world to measure a horse’s mouth length in order to choose an appropriately-sized bit (the metal piece that fits in a horse’s mouth) for riding. Bit length is important because a too-small bit can pinch or damage the skin of the mouth, and a too-large bit can slide through the horse’s mouth and knock against cheek teeth while riding. In an archaeological context, we can take an actual historical bit and compare it to our baseline collection of mouth/bit lengths from modern horses. Together with other data, bit length could give us an idea of the potential size of the ancient horses who may have worn them.

A standard height-weight horse and pony tape is also used to calculate the height of each horse, which is important allometric data along with general information, such as the horse’s known breed (which indicates its body type, such as pony, small stock horse, draft, or warmblood). Even if a horse’s height is known by the owner, it’s good scientific practice to maintain consistency with data recording; one person takes each measurement for each horse, therefore limiting the human error factor.

While actual hooves have a small chance of appearing in the archaeological record, horse shoes are much more durable. And because horses wear many, many sets of shoes in their lifetime, it is more likely that we might find historical horse shoes. But what good are these artefacts if we have nothing to compare them to? How do we know the size of the horses who wore them? Taking a look at the size and shape of many living horses’ feet can be one way to figure that out. Measuring something as variable and oddly shaped as a horse’s foot can be tricky though, and this is where a little bit of creativity is needed. We have called the device developed for Warhorse our “Hoofometer”; it is a simple piece of clear plaskolite with two traditional scale bars added. By holding a horse’s hoof (as one does when cleaning their feet) and holding the hoofometer flat against the bottom of their hoof, it is possible to photograph it and therefore create a picture that can be shared with other researchers. They can then choose which measurements to take and how to define the shape and size of the foot.

Another tricky measurement that requires a creative solution is the hip of the horse. Rounded and bulging with muscle, it is not the kind of body part that can be measured with a simple seamstress tape. Such a measurement would result in a contour that follows the slope of muscle, and a contour measurement is a lot harder to relate directly to a skeletal measurement. Two metre sticks and some light hardware and scale bars are all that is needed to create a device that takes straight measurements of large muscled areas such as the hip. The sliding scale has two arms that rest against the chosen data points, resulting in a number that is a little closer to the greatest length of the pelvis.

Modern methods (such as the use of bit sizers for bit length and measuring a horse’s hoof for shoe size) can sometimes be useful when examining material culture in an equine context, and while not directly comparable, live body measurements can provide an interesting dataset next to skeletal measurements. Zooarchaeology can sometimes require interesting experiments and creative thinking, which makes it an endlessly enjoyable endeavour. Plus, any data-gathering that requires paying the subjects in carrots is a special bonus! I am immensely grateful to the team of Warhorse for letting me join their fun, and I look forward to the results of this great collaborative project.

Digital Camera


Dr Gary Baker

A few months ago I wrote a blog looking at how England’s stock of medieval warhorses were managed: MANAGING ENGLAND’S ROYAL WARHORSES IN THE FOURTEENTH CENTURY | Warhorse ( This month I look in more detail at how English monarchs kept their horses provisioned, and the administrative machinery behind that process.

An army marches on its stomach, so the saying goes, and what is true of people is equally applicable to animals. Medieval horses, just like their present-day counterparts, were used for a variety of different tasks. Indeed, in the Middle Ages the use of horses was far more widespread than it is today, with horses employed in a whole multitude of tasks from agricultural work to war. Whatever their intended function, horses require a large number of calories to stay healthy and perform their duties. A typical horse eats between 2­–2.5% of their bodyweight as dry forage per day, but this varies depending on age and workload. A modern horse undertaking ‘heavy work’ such as racing will need somewhere in the region of 30,000 calories per day. Clearly this is at the top end of nutritional requirements for horses bred for peak physical performance. Nevertheless, even horses utilised in less strenuous physical activities require a substantial amount of fodder. The majority of this can be met through grazing for horses ‘at rest’ or employed in light work, but grazing alone cannot meet all of the calorific and nutritional needs for the majority of horses.

The accounts of England’s royal warhorses provide a huge corpus of information about the provisioning of the king’s elite horses. The specifics of each royal stud and stable vary based upon location, particularly the availability, and quality, of pasture for grazing. Nevertheless, in general terms, we are able to detect patterns in the accounts. Many, if not all of the royal studs were within, or adjacent to, royal parks and manors, and the horses were pastured and grazed within them during the spring and summer, with additional sustenance provided by royal officials. In the colder winter months however, these officials – the keepers of the studs and stables and the grooms they employed to maintain the animals – provided the horses with the majority of their food. The major fodder provided to the animals were oats, hay, straw, and cut grass, with the occasional mention of other foodstuffs like bran, beans, peas, and ‘horse bread’, a mixture of grains and ground legumes. The amount provided for each animal depended on its age and type, but ascertaining exactly how much is tricky. The amounts of hay, cut grass, straw, and horse bread, are not provided in the accounts, only the expense of buying them. Even the more expensive foodstuffs, the quantities of which are provided in the accounts, are difficult to ascertain with certainty, as they are measured in quarters, bushels, and other often unfamiliar terms. The bushel, for example, was a measure of volume varying in weight from less than 25lbs to as much as 50lbs, and 8 bushels comprised 1 quarter.

Nevertheless it is possible to provide approximations of horses intakes from the information in the documents. Unsurprisingly it was the destriers, the elite male warhorses, which received the lion’s share of the food. At a conservative estimate this amounted to a half to two-thirds of a bushel of oats per day (c.12lbs per day with the half bushel), but this could be as much as two bushels, such as was provided for 3 stallions at Eastwood in October 1294. This meant that if fed only a half bushel every day, a destrier would consume 91¼ bushels (or just under 11½ quarters at 8 bushels the quarter) per year. This was far more than the average medieval carthorse at around 0.2 bushels per day, and triple the amount of the average medieval plough horse at c.1–4 quarters per year. It has thus been calculated that if destriers were also receiving more hay and other foodstuffs than other horses, then the destrier may well have been as much as 400–500lbs heavier than the average medieval horse. Colts received considerably less in oats per day than destriers, whilst mares and fillies seem to have received no oats at all, instead being fed largely on hay and straw.

An illuminated manuscript from the German ‘Sachsenspiegel’ law code (early 14th century). Horses used for agriculture required far less food than warhorses.

Large quantities of these foodstuffs had to be purchased for each animal at considerable expense. The account of Arnold Garcy, Keeper of the king’s ‘great horses’ (a catch-all term applied to a variety of different types of animal) from 26 October 1330 to 15 December 1331 is illustrative of the level of this expense. Arnold’s account records a total expense for the account of £783 3s 8½d for as many as 112 horses: destriers, coursers, hobbies, and Arnold’s own horse. Of this amount, a total of £524 2s 6¼d was spent directly on food for the horses: £145 19s 1½d on hay; £352 16s 8¼d on 1561 quarters and 3 pecks of oats and 46 quarters and 5½ bushels of bran and an unspecified amount of horse bread; £11 15s 10d of cut grass for the horses in the summertime; and £13 10s 10½d on straw. In other words, two thirds of all Arnold’s expenses for the period were for feeding the horses under his care.


Detail from the account of Arnold Garcy, Keeper of the King’s Great Horses, 26 October 1330 – 15 December 1331 (TNA E372/176 f. 63v).

These foodstuffs were acquired from a variety of sources local to the stables for which they were required, though sometimes they were brought via water, especially to those studs and stables in Oxfordshire from which large quantities of foodstuffs were shipped up the River Thames from London. The scale of the logistical operation of supplying a royal stable can be mapped thanks to rolls of provisions compiled by the keepers of horses, showing to whom they still owed money for the fodder they had been supplied. The map below shows the provisioning network for a single royal stable in the village of Eynsham in West Oxfordshire, from 12 January – 6 April 1344, with the locations which supplied the stable with oats (purple), beans and peas (green), straw (pink), and hay (black).

The map shows just how extensive these provisioning networks were, though more research is needed into whether or not the presence of the king’s horses in a stable was a burden on the locality, draining it of much needed foodstuffs, or a boon, providing income and jobs for those in the locality.