Archaeological Excavations at the North Annexe, York Guildhall

Mammal and Bird Bone | Fish Remains | Avian Eggshell | Mollusc Remains | Plant Macrofossils and Charcoal

7.1 Mammal and Bird Bone by Kris Poole

Introduction

This section covers the final analysis of the mammal and bird remains recovery during excavations at York Guildhall. The earlier assessment bone report (Poole 2022) recommended that only bones from Romano-British (Phases 3, 5, 7–14 and 16) be subject to final analysis, as material from later contexts was particularly subject to issues of residuality and reworking. A total of 431 fragments of animal bone recovered by hand collection and 4,956 retrieved from environmental residues (the vast majority of which were tiny, unidentified fragments) were recorded for this final report. This excludes fish remains, which have been studied by Hannah Russ and are discussed elsewhere (see section 8). Although the stratigraphic phasing is referred to here, due to the small size of the assemblage, material is also grouped into the four periods set out by the pottery discussion (see section 7.1): Period 2 (Phases 3–5), Period 3 (Phases 6–9), Period 4 (Phases 10–13) and Period 5 (Phases 14–17). The largest quantity of bone belongs to Period 4, dating to the mid-2nd to early 3rd century CE, with smaller amounts of bone from all of the other broad periods.

Methods

Levels of preservation were recorded using Behrensmeyer's (1978) standards, with burning and gnawing also recorded. Butchery was recorded in detail, noting the butchery mark type (chop, cut, saw, shave) and its location on the bone. Attempts were made to identify all bone fragments to element and species, with some exceptions. Mammal ribs, vertebrae, skull fragments and long bone fragments not identifiable to species were classed as large-, medium-, or small-sized mammal (except for atlas and axis vertebrae, and the more durable/diagnostic parts of the cranium, namely the zygomatic, occipital, maxilla and horn core, which were identified to species). Ribs were only counted when the rib head was present. Apart from the calcanei and astragali, carpals and tarsals were not recorded. Similarly, for birds, all elements were identified, where possible, to species, apart from vertebrae and ribs, which were classed simply as 'chicken size' or 'goose size'.

Morphological criteria of Boessneck (1969), Payne (1985), Prummel and Frisch (1986) and Halstead et al. (2002) were utilised to attempt to distinguish between sheep (Ovis) and goat (Capra). Red deer was distinguished from cattle using Prummel (1988). Presence or absence of a pneumatised proximal foramen of the femur and the continuation of the medial calcaneal ridge on the tarsometatarsus were used to distinguish chicken/guinea fowl from pheasant bones (Cohen and Serjeantson 1996, 63, 79), although no bones were positively identified as pheasant or guinea fowl. Geese lack suitable morphological criteria on which to differentiate between individual species, and there is also considerable size overlap between species (Barnes et al. 2000, 91). The domestication status of geese and ducks in the Roman period is also unclear. Accordingly, goose bones were recorded as Anser/Branta (although they were compatible in size with domestic goose), whilst the single duck bone was recorded as 'mallard-size'.

All identified fragments were recorded as individual specimens, with the exception of fresh breaks which were refitted where possible and counted as one element. Partial or complete skeletons were recorded as one specimen, with details of the elements present, completeness and measurements noted.

The zoning systems set out by Serjeantson (1996) for mammals and Cohen and Serjeantson (1996) for birds was used to record elements. These were used to work out the Minimum Number of Elements (MNE) for cattle, sheep and pigs.

Grant's methods (1982) were used for recording tooth wear in cattle, sheep and pigs, with age categories assigned following O'Connor (2003, 158–65), to enable comparison with other York assemblages. Epiphyses were recorded as 'foetal', 'neonatal', 'unfused', 'fusing' or 'fused'. Bone fusion data for sheep/goat, cattle and pigs were grouped into 'early', 'intermediate' or 'late' fusing epiphyses, following O'Connor (2003, 166). As bird bones lack epiphyses, elements were recorded as 'fused' or 'unfused.'

Morphological and metrical traits of the pelvis and horn cores were used to sex sheep/goat (Hatting 1995; Greenfield 2006), although no suitable landmarks were present to enable sexing of cattle or pigs. Presence or absence of tarsometatarsi cockspurs and the presence of medullary bone was used to differentiate between male and female chickens (Driver 1982).

Measurements were taken following von den Driesch (1976) for mammals and Cohen and Serjeantson (1996) for birds. Withers (shoulder) heights were calculated using von den Dresich and Boessneck (1974), except for cattle astragali, for which Tsalkin (1970) was used.

Results

Taphonomy

All of the remains derived from a series of layers and deposits associated with demolition and dumping of material on the site, with the exception of Phase P10 and P11 (surfaces and associated deposits) and P14 (wall foundation cut). As such, it can be assumed that most of the material accumulated elsewhere, perhaps in midden deposits, before being dumped on the site, in part to raise the ground level. Each of these deposits were, however, variable in the species/elements they contained, some containing bones from across the body, with others comprising more head and/or foot elements, whilst species proportions varied (see butchery and body-part patterns below).

Bone condition was only assessed for hand-collected remains (due to the small size of the bones from residues), and the vast majority of the bones were determined to be in fair condition, with smaller numbers in good condition. Despite the potential accumulation of this waste prior to dumping, levels of gnawing on the hand-collected bones were relatively low, with an overall average of 5.8% (Table 34). None of the Period 2 bones were gnawed and only one of the Period 3 bones had gnawing, but the sample sizes for those contexts were the smallest of any period. The material was not unduly fragmented, with proportions of loose teeth for the hand-collected bone at just 0.03% of the assemblage. Only a single bone, a cattle radius, had traces of burning amongst the hand-collected material. Burning was more in evidence in the bones from residues, although still made up only approximately 5% of the bones.

Species identified

As would be expected of a Romano-British urban context, domestic species and particularly the three main domesticates (cattle, sheep/goat and pig) dominated the assemblage. When suitable landmarks were present on bones of sheep/goat, a small number could be identified specifically as sheep, with none being determined to derive from goats (Tables 35 and 36). Accordingly, the term 'sheep' is used to cover all ovicaprid remains. Notably, although cattle are the most frequently represented species in the hand-collected bone, pig was the second-best represented species, followed by sheep. This holds true for the assemblage as a whole and Period 4 material, but not the other periods, although sample size may be a factor.

Differences in the proportion of larger mammals (cattle) as opposed to medium-sized mammals (i.e. sheep, pigs) is evident between the hand-collected and residue assemblages, with cattle bones overall being proportionately more frequent in the former (cattle 38%, sheep 29.6%, pig 32.4%) than the latter (cattle 21.5%, sheep 44.3%, pig 34.1%). The same broad pattern holds true when considering Period 4 bones alone, although the relative proportions of sheep against pig are different in the residues (Hand collected: cattle 38.1%, sheep 25.7%, pig 36.2%; Residues: cattle 17.1%, sheep 44.3%, pig 34.1%). This difference between cattle proportions by recovery technique is a relatively common phenomenon for archaeological bone assemblages, with hand collection tending to over-represent larger species such as cattle and under-represent smaller species. Nonetheless, under both recovery techniques, sheep and pig bones together are more common than those of cattle. The other domestic species in the assemblage, dog and horse, were represented only by single bones.

Wild mammals were present in the form of a single red deer radius from a P16 context, C8069, in addition to small mammal bones, frequently partial skeletons, in the residues. Most of the smaller mammal bones lacked diagnostic elements, although the presence of house mouse was established for P16 (Period 5) context C19161, and other bones were from either mouse or vole.

In common with the hand-collected assemblage, most bird remains were from domestic species. The bones of Galliformes (chicken/pheasant/guinea fowl) were the most frequently represented bird remains. Although feasibly some of these remains could derive from pheasants or guinea fowl, no evidence of guinea fowl has to date been identified for Roman Britain, and for pheasants, few positive identifications have been made (Poole 2010). It is assumed that all of the galliform bones are from chickens, but the possibility of other species, even if small, needs to be borne in mind. The few instances of goose and duck bones were consistent with domestic species, although the domestic status of these species in the Roman period is uncertain (Albarella 2005).

Of the four wild bird bones, corvids were represented by three bones, one from a raven and two from crow/rook (all from P12 in Period 4). The fourth bone was particularly interesting, coming from a common crane (Grus grus). This was recovered from a sample from C19122 in P13, the earliest record of this species from within York (see Discussion).

Ageing and Sexing

The mandibular ageing data showed an interesting contrast between mandibles of cattle with those of sheep and pigs (Table 37). Three cattle mandibles could be aged, from three different periods. All were from adult individuals and judging by the advanced wear of the lower third molar of some these individuals were at quite an advanced age. By contrast, irrespective of period, none of the sheep or pig mandibles were from adult individuals, with sheep mandibles being either immature or subadults and pigs being juvenile, immature or subadult in age. Even so, it must be acknowledged that the numbers of mandibles are very low.

Epiphyseal fusion data were calculated only for Period 4 bones. The numbers indicate that, for cattle, the majority of adults were present in the assemblage, although they do also suggest the presence of some immature animals, at the intermediate and late fusing stages. In contrast to the mandibular ageing data, epiphyseal fusion does also indicate that at least some sheep in the assemblage made it to maturity. However, in line with the dental ageing, none of the pig epiphyses were from animals which had reached maturity, with only some of the earliest-fusing elements having fused, suggesting that no pigs in the assemblage were adult at death.

Sexing information for mammals was limited to two sheep pelves and a horn core, which were all from males.

All of the bird bones in the assemblage had fused. Sexing information was limited, with no complete tarsometatarsi, meaning that the presence or absence of cock spurs could not be assessed. Where bones could be assessed for the presence of medullary bone (indicating hens in lay), this was not present. However, this may simply indicate birds being killed when not in lay, rather than the presence of males in the assemblage. Similarly, none of the three goose bones in the assemblage contained medullary bone.

Butchery and Body-Part Patterns

Table 38 sets out the Minimum Number of Elements (MNE) for cattle, sheep and pigs, based on Period 4 contexts. For all of these species, it can be seen that all parts of the body are represented to some degree, with head (horn core, maxilla, mandible), upper limb (scapula, humerus, pelvis, femur), lower limb (radius, ulna, tibia) and feet (metacarpals, metatarsals, phalanges). This does not mean that all elements were found in the same contexts. The largest deposit in terms of bones recovered was C3026 from P13; for this context, the pattern was similar to that for the Period 4 contexts as a whole (Table 39). Other contexts contained a much smaller number of bones, such as C19137, and so are harder to interpret (Table 40). In this context, most bones were from cattle and included head, upper limbs and foot elements, suggesting a mixture of some primary and secondary butchery waste, but sheep was only represented by two bones and pig by one bone. What C3026 and C19137 have in common, however, is the presence of bird remains. Processing of bird carcasses is typically a kitchen or table activity, so direct consumption waste appears to be present in each case. Of the 19 contexts from Period 4, all contained at least one bone from birds typically eaten in the Roman period, suggesting that consumption waste was present throughout these deposits.

A total of 34 bones had signs of butchery, of which 15 were from cattle, seven from large mammals (probably cattle), four from sheep, three from pigs, two from medium mammals, one from red deer, and one from chicken, making up c. 6% of bone recovered by hand collection. The types of butchery consisted of 14 chop marks (using a cleaver or similar tool), 17 cut marks (from knives) and three bones with shaving marks.

Metrics

Withers (shoulder) heights were calculated for a cattle metacarpal (105cm) and two astragali (102cm, 115.7cm), falling either side of the average heights for cattle in the 24–30 Tanner Row assemblage (O'Connor 1988) and 3rd century CE deposits from Lincoln (Dobney et al. 1996). Most of the limited number of cattle measurements were width and depth measurements, and these fit within the range seen at Tanner Row and Lincoln. Sheep withers heights were: 57.2cm and 57.4cm (calculated from astragali), 104.4cm and 115cm (based on metacarpals) and 113.5cm (a metatarsal). In general, sheep measurements from this site tend to be lower than for other sites in York and in Lincoln. A single pig astragalus gave a withers height value of 58cm.

Discussion

As discussed above, the material brought onto the site is likely to represent, at least in part, material that accumulated elsewhere in the town, perhaps as middens, before being dumped. Given the proximity of the site to the legionary fortress, it is possible that at least some of the deposits derived from within the fortress itself, or the extra-mural settlement that formed around its perimeter. Support for this is perhaps provided by the pottery recovered from these contexts, which included high proportions of fine wares (see Fine Wares). The exact sources of the material deposited at the Guildhall cannot be definitively traced, but analysis of the bones suggests that a range of different carcass processing events are represented in the assemblage as a whole, but also within single contexts. These differences should not be overstated, given the low numbers of bones, but do hint at a diversity of origins for the material ultimately dumped at the Guildhall site.

Taking the assemblage as a whole, what is perhaps most notable is the lack of large dumps of cattle bones, or intensively butchered bones, which would represent similar systematic butchery of cattle carcasses as seen at other sites in York, such as at 24–30 Tanner Row (O'Connor 1988), and which are also common at other Roman urban centres, including Lincoln and Exeter (Dobney et al. 1996; Maltby 1979a; 1979b). This is perhaps one factor that has led to pig bones being almost as frequent as cattle bones in the hand-collected component, and that together, sheep and pig bones outnumber those of cattle. The pattern is even more marked when bones from residues are considered, and these compare with the large assemblage from 24–30 Tanner Row (O'Connor 1988), and smaller assemblage from Queen Street (Poole 2023), both within the colonia, and a small assemblage from Hungate (Rainsford nd), in which cattle dominated. More generally, cattle are typically the most commonly represented species on military sites and in large towns (Maltby 2014).

The high proportion of pigs in the Guildhall assemblage is also noteworthy, being the second most commonly represented species in the hand collected remains (after cattle) and in the residues (after sheep). In most Romano-British bone assemblages, pigs are third behind cattle and sheep, although they are generally found in higher proportions within major towns, military sites and some villas compared to small towns and rural settlements (Maltby 2014, 797). Pigs are also frequently better represented in assemblages recovered from sites more centrally placed within Roman towns, near the forum and basilica, than in their suburbs (Maltby 2014, 797). The Guildhall assemblage would fit with this pattern. This distinction may reflect dietary differences, with high demand for pork amongst those of higher status in towns (high percentages of pigs often being found at high status sites). It could also reflect cattle carcasses being processed away from the central areas of towns, and thus the waste from such activities did not inflate cattle bone frequencies in assemblages from more central locations.

A number of bird bones were recovered from the site, with chicken being most commonly recovered. At 11.6% of the hand-collected assemblage (compared to cattle, sheep and pigs), this proportion is higher than the average (5.6%) for major Roman towns, which themselves typically exhibit the greater proportions of chicken remains than other Roman site types (Maltby 2018). When comparing only sheep and chicken bones, the proportion of the latter is 30.7%, also higher than for major towns generally (at 19.2%). Presence of chicken remains in all Period 3 contexts suggests they may have been a relatively frequent dietary component.

In addition to the chicken remains, four bones were from wild species (as noted above, the geese and duck bones could potentially be domestic or wild). A diverse range of wild bird species have been recovered from Roman Britain, although their remains generally only make up a very small proportion of bone assemblages (Serjeantson 2023, 173–4; Yalden and Albarella 2009, 108), and that is the case for the Guildhall assemblage.

Of these species, corvids (crow family) are one of the most commonly found wild birds that have been identified in Roman assemblages (Serjeantson 2023, 21; Yalden and Albarella 2009, 108). Whilst corvids are known to have symbolic importance in the Roman world, the incorporation of raven and crow/rook remains with the other remains at the Guildhall site probably has a more prosaic explanation. As broad-spectrum omnivores, and known scavengers, ravens and crows/rooks would have exploited the concentration of food and waste provided within an urban setting (O'Connor 1993). Corvids have been identified in bone assemblages from York from the Roman period onwards and are the most regularly encountered species after domestic ducks, geese and chickens (O'Connor 2013, 114).

In contrast to corvids, remains of common crane are less commonly found on Romano-British sites. Surveys of bird remains from archaeological sites indicate that bones of the common crane have been recovered from approximately 10–14% of bone assemblages dating to the Roman period (Allen 2011; Serjeantson 2023; Yalden and Albarella 2009). They are found on a range of different site types, including urban and military, but also some lower-status rural settlements. Evidence from other sites (such as Fishbourne, Carlisle and Caerleon), in the form of cut marks, indicates that crane was eaten and may have been prized – butchery on a crane skull from Caerleon was comparable with preparation techniques recommended by Roman writers (Yalden and Albarella 2009, 111). How applicable such texts are to attitudes towards cranes across Roman Britain are, however, uncertain. What is clearer is that the crane bone from the Guildhall site represents the earliest known example of this species from York. Another example was found in 4th century CE deposits at 24–30 Tanner Row (O'Connor 1988), whilst examples are also known from Anglo-Scandinavian and later medieval deposits (Bond and O'Connor 1999; Claire Rainsford, pers. comm.). There is no definitive evidence of falconry being practiced in Roman Britain (Poole 2018), indicating that this individual would have been caught by other wildfowling methods, such as nets or snares; such methods are recorded as being used by officers at Vindolanda to catch thrushes, ducks and swans (Serjeantson 2023, 5).

The single red deer bone in a 4th century CE context provides some limited evidence of hunting, and the cut marks on this bone indicate that it represents consumption waste. Deer remains are only ever present in very low numbers on Romano-British sites, with red and roe deer generally constituting less than 1% of food mammals on urban sites (Maltby 2014, 799). Deer hunting was potentially a status marker during this period (Allen, M. 2014), although deer bones can appear on a range of site types.

Conclusion

The Guildhall assemblage has provided useful information regarding diet and activity within central York during the Roman period. The bones represent the remains of a mixture of primary and secondary butchery, as well as kitchen and table waste. Despite the presence of some butchery waste, the assemblage did not indicate the large-scale processing of cattle carcasses seen elsewhere in York. Species frequencies indicate a high proportion of pigs, perhaps suggestive of status-based dietary patterns, potentially due to the proximity of the fortress and wealthier households in this central area. Bones from crane and red deer could also reflect status markers, linking with the pottery assemblage in its high number of fine wares. All of these interpretations need to be treated with some caution, given the small size of the assemblage, but should further bone assemblages of Romano-British date be recovered from this area of the town, it will be possible to refine the suggestions outlined here.

7.2 Fish Remains by Hannah Russ

Introduction

Fish remains (147 fragments) were recovered from bulk environmental samples taken during archaeological excavations at the Guildhall Site.

Methods

Fish remains were identified to element, side and the lowest taxonomic level possible using the author's reference collection and identification guides (Archaeological Fish Resource; Camphuysen and Henderson 2017; NABO nd; Osteobase; Wheeler and Jones 1989). Quantification used the diagnostic zone method as presented by Barrett (2001) and Harland et al. (2003). Remains of the cod order fish (Gadiformes) were allocated to size categories as described by Cerón-Carrasco (2004). Where possible other fish remains were allocated to the same size categories based on comparison with reference specimens of known total length.

A taphonomic assessment of each fragment was undertaken, recording the presence and absence of cut (knife) and chop marks, burning and calcination, any evidence for animal activity (canid or rodent gnawing), pathology and surface preservation; any other surface modifications of note were also recorded. Fragments of bones that could be identified to element but not any specific species were grouped as far as possible using size and class or order categories. Results were recorded in an electronic proforma in Microsoft Excel.

This analysis has been undertaken in line with current standards and guidelines (CIfA 2014; Baker and Worley 2019).

Results

Fish remains (n=147) were recovered from bulk environmental samples from 13 contexts across the site, from features and deposits associated with activity Phases 5 to 20, and Boreholes 3 and 4 (Table 41).

The fish bone assemblage included the remains of marine (n=30) migratory (n=32) and freshwater (n=32) species and was recovered from Roman period deposits and features, dating from the late 1st to 3rd centuries CE, and medieval deposits and features dating to between the late 11th and early 13th centuries, Table 41. It is suggested that some fish remains from the medieval deposits are residual from Roman activity at the site based on the presence of Atlantic chub/Spanish mackerel being present in medieval deposits, this being a Roman import into the city that would not be expected in the medieval period.

Roman period: Late 1st to 3rd century CE (Phases 5 to 13)

Fifty-nine fish bones were recovered from ten Roman period contexts across the site (Phases 5 to 13), Table 41. All of the specimens were recovered from bulk environmental samples, without which evidence for fish consumption at the site would have been absent. Most of the Roman period fish remains came from Phase 11 deposits (n=34) representing a structure and associated activity located at the south-western end of Excavation Area 2 dating to the mid-2nd to early 3rd century CE. Small numbers of fish remains were recovered from Roman Phases 5 (n=6), 7 (n=1), 9 (n=6) and 13 (n=12) (Table 41).

The Roman period assemblage is dominated by the remains of small and tiny fish (up to 30cm total length but mostly 15-20cm) of the carp family (Cyprinidae), forming 52.9% of the identified fish remains (n=27). It is very difficult to attempt species identification for cyprinid fish based on comparative skeletal morphology for most elements; pharyngeal bones are an exception to this and two pharyngeals from C19188 were identified as European chub (Squalius cephalus). The only other freshwater species recorded was a single vertebra from a small pike from C19176. The next most frequently occurring species was the migratory European eel (Anguilla anguilla), which formed 27.5% of the identified portion of the fish remains from Roman period contexts. Marine fish were rare but included two clupeid species, Atlantic herring (Clupea harengus) and European Sprat (Sprattus sprattus) as well as three vertebrae from Atlantic chub/Spanish mackerel (Scomber colias). The species recorded at the Guildhall are consistent with those seen elsewhere in York (O'Connor 1988; Jones 1988; Enghoff 2000) and across Britain (Locker 2007), with the dominance of cyprinids followed by eel also seen in the assemblage from Roman (Period 3–7) deposits at Tanner Row in York (O'Connor 1988, 114). Elsewhere in Britain, cyprinids are less frequently recovered, with eel, herring and plaice/flounder most often forming the three most abundant species (Locker 2007, 146). The abundance of cyprinids at York is likely a result of its location close to the River Ouse where these freshwater species could have been caught.

Of note were vertebra from Atlantic chub/Spanish mackerel (Scomber colias Gmelin 1789) from Phases 5 and 11, as well as one from medieval Phase 19, which is considered to be residual in that deposit from earlier (Roman period) activity at the site. The Spanish mackerel is not native to the seas around Britain, but, amongst other areas, lives in the Atlantic Ocean around the coast of Spain and in the Mediterranean Sea. Historically, in the archaeological literature, this species has been referred to as Scomber japonicus Houttuyn 1782, a name now used to describe the Indo-Pacific mackerel (Harland 2018, 427; van Neer et al. 2010). The species has been recorded in small numbers at other Roman sites in Britain, including at Vine Street (Nicholson 2009; Browning et al. 2021) and Waterside (Russ and Chorro-Giner 2022) in Leicester, Chester Amphitheatre (Harland 2018, 427) and 25 Bridge Street, Chester (Jaques et al. 2004), Gorhambury villa, near St. Albans (Locker 1990), Great Holts Farm, Boreham, Essex (Murphy et al. 2000) and London, Winchester Palace (Locker 2007; Yule 1989; 2005). There has also been a single find in a Late Iron Age context at the oppidum at Skeleton Green, Stevenage (Wheeler 1981). These bones represent the remains of preserved whole or sections of fish, a product known as salsamenta, which was primarily made from mackerel, tuna and/or scad (Harland 2018 428) and transported from the Mediterranean across the Roman Empire in amphorae (van Neer et al. 2010). The Guildhall finds from Phase 11, along with the Leicester Vine Street and Waterside examples are of a later date (2nd–3rd century CE) than most British finds, which tend to date towards the end of the 1st century CE (Harland 2018, 429; van Neer et al. 2010, 168).

Medieval period: Late 11th to early 13th centuries (Phases 18–20)

Eighty-one fish bones were recovered from medieval phases of activity at the site dating to between the late 11th and early 13th centuries (Table 41). Eel (n=16) and herring (n=12) were the most frequently identified species. The fish remains from these deposits are unusual for those recovered from contemporary deposits in Britain; the species represented are similar to those seen in the Roman period, and with the presence of Atlantic chub/Spanish mackerel, which represents a Roman import into York, it is suggested that many of the fish remains recovered from medieval features and deposits are residual from Roman activity at the site. Possible exceptions might include the gurnard (Triglidae) and ray (Rajidae) bones from C19075 which, while not unheard of in Roman period assemblages, are more common in those associated with medieval activity.

Whether the fish remains from medieval features are residual or not, they are still inconsistent for what would be expected at a medieval city such as York. During the 12th to mid-13th century fish, especially marine fish, start to play an increasing role in the diet of people in Britain (Serjeantson and Woolgar 2006, 111), with Atlantic herring being the most frequently recovered species, probably arriving in towns and cities salted or pickled in barrels as this species spoils quickly unless preserved in some way. The availability of preserved herring increased through the medieval period such that by the later medieval period they were a widely available and cheap source of nourishment available to all but the very poorest of society (e.g. Barrett and Orton 2016; Serjeantson and Woolgar 2006 116), packed in barrels in their millions and distributed from the coasts throughout England. From the 11th century in Britain, stockfish became a popular and widely available resource, favoured for their long-term preservation (Locker 2000; Barrett et al. 2004; 2008; 2011; Serjeantson and Woolgar 2006; Orton et al. 2011). Stockfish was often produced from large Atlantic cod, but ling and conger eel were also used. Freshwater fish were also increasingly consumed during the medieval period (Locker 2018; Hammond 2005, 22). The eel, a migratory fish, was also an important food species as well as serving as a form of currency in England, used/required until the 17th century (Greenlee 2020). Eels were so popular in the medieval period that their popularity and the demand for them saw them being imported to England from Belgium (Woolgar 2000, 36). While the technological advances in fishing and fish preservation no doubt increased the availability of fish, Christianity also played a role in increasing fish consumption through the enforcement of 'fasting' days where meat consumption was prohibited (e.g., Kurlansky 1999; Woolgar 2000; Fagan 2006).

Given the above, it would be expected that fish remains would be more abundant in medieval deposits, and include more marine species, especially Atlantic cod and other cod species. Therefore, the absence of these remains suggests that the excavation areas did not include features that were used for the discard of food waste, though this suggestion is not supported by the mammal and bird bone data and marine mollusc data (see section 7.1 and section 7.4). The only remaining explanation is that, for reasons unknown, the increased consumption of fish that is evidenced across Britain at this time did not occur at, or in the vicinity of, the Guildhall.

Conclusion

The small Roman period assemblage demonstrates that marine fish caught from inshore coastal or estuarine areas as well as preserved fish brought from the Mediterranean region played a small role in the diet of those living in and visiting the site at Guildhall during the early to mid-Roman period (1st to 3rd century CE). The recovery of remains from fish caught off the coast around Britain demonstrates the presence of an established trade network that could deliver fish to inland locations before they spoiled. However, freshwater fish of the carp family and the migratory European eel were more important. With the exception of the Spanish mackerel bones, which represented larger fish, all of the bones were from tiny to small-sized fish and could represent remains of fish products made locally in the Mediterranean style used for producing garum and other fish sauces.

The Spanish mackerel bone contributes to the small but growing volume of evidence for the importation of salsamenta into Britain during the Roman period, and together with the vertebra from Waterside (Russ and Chorro Giner 2022) and Vine Street in Leicester (Nicholson 1999: Browning et al. 2021), and remains from Great Holt Farm, Boreham, Essex (Murphy et al. 2000), suggest that this trade was not a practice only seen in the 1st century CE, but one that continued into the later Roman period.

The remains recovered from medieval deposits seem to be, at least partly, residual from Roman activity. Either way, it is clear that while there is medieval activity at the site, fish remains consistent with medieval fish consumption practices were not disposed of in the areas excavated at the Guildhall.

7.3 Avian Eggshell by Hannah Russ

Introduction

Fragmentary remains of birds' eggs weighing 1.7g were recovered from three bulk environmental samples, all from Roman period deposits. This section includes quantification of the assemblage, and discussion of the findings in the context previous work on bird eggshell identification; methods for identifying species are also discussed.

Methodology

The remains were counted and weighed by context to the nearest 0.1g. This work has been undertaken in line with published standards and guidelines (CIfA 2014; Campbell et al. 2011), with reference to the Yorkshire Archaeological Research Framework (Roskams and Whyman 2005; 2007).

Results

The small assemblage was recovered from the 2–4mm or 2–8 mm fractions of the heavy residues/retent from bulk environmental samples 65, 68 and 69 (Table 42). Two contexts, C19176 and C19188, were occupation deposits overlying a Period 4 floor surface, dated by pottery to the late 2nd to early 3rd century CE. The third, context C19181, was part of a Period 6 levelling deposit containing late 12th to early 13th century pottery; however a large quantity of residual Roman material was also present.

Discussion

It is not possible to identify bird species from eggshell without the use of scanning electron microscopy (Kippax 1981; Sidell 1993), ZooMS (Stewart et al. 2013; 2014) or other protein peptide analyses (Presslee et al. 2018). Bird remains at the site might point towards chicken, goose and/or duck. It is also possible that wild bird eggs were collected for consumption. Though later in date than those recovered at the Guildhall, analysis of eggshell recovered from early medieval contexts at Coppergate and Hungate in York identified most remains as chicken eggshell, with a small contribution from goose and duck (Stewart et al. 2014). The quantities of eggshell recovered suggests that eggs played at least some role in the diet of those living at or in the vicinity of the Guildhall during the 2nd or early 3rd century CE.

There has been very little scientific analysis of eggshell recovered from Roman period sites in Britain, and none that has used ZooMS or other protein peptide analyses, as such, the assemblage is an important resource for future research in eggshell identification and Roman period diet and subsistence strategies in York.

7.4 Mollusc Remains by Hannah Russ with assistance from Emma Tong

Introduction

Mollusc remains (624 fragments weighing 5.37kg) were recovered via hand collection and from bulk environmental samples. The mollusc assemblage included the remains of marine (n=616) and terrestrial (n=8) species and was recovered from deposits and features dating between the 2nd century CE to modern periods.

Methodology

The mollusc remains were identified to side where appropriate, and to as low a taxonomic level as possible using the author's reference collection and published and online identification guides (Cameron 2003; Hayward and Ryland 1995; Kerney and Cameron 1979; Naggs et al. 2014; Pfleger 2000). Quantification used a diagnostic zone method, measurements were taken according to Claassen (1998, 109–10) with European flat oysters (Ostrea edulis) recorded according to Winder (2011).

Oyster size was estimated for near-complete specimens that did not qualify for measurement using a scale from very small to very large. A taphonomic assessment of each fragment was undertaken, recording the presence and absence of any infestations and evidence for tool marks or burning; any other surface modifications of note were also recorded. Results were recorded in an electronic proforma in Microsoft Excel.

This work has been undertaken in line with published standards and guidelines (CIfA 2014; Campbell et al. 2011; Campbell 2015; Campbell 2017; Winder 2011),

Results

Terrestrial Molluscs

Terrestrial mollusc remains were recovered in very small numbers from seven contexts (Table 43). The identified remains included the common garden snail (Cornu aspersum) and members of the glass snail family, Oxychilidae. In some cases, the remains of terrestrial snails can inform on the nature of past environmental conditions at a given location. However, the snails identified provided limited information in this regard. Many of the glass snail species are considered 'shade-loving' (Evans 1972), but none of the remains were identified to species so while their presence gives a broad indication of possible past environments, the absence of a species identification and their rare occurrence precludes any definitive or specific comments. The common garden snail is considered a catholic species (Evans 1972), occurring in such a wide range of habitats that it is not useful in indicating any specific past environmental conditions. The small assemblage is also too small to provide any meaningful information; it spans a period of over 1000 years, reducing the sample size to just a few specimens per chronological period (Table 43).

Marine Molluscs

The remains of marine molluscs (616 fragments weighing 5.37kg) were recovered from 102 contexts during the excavations (Table 44). The remains were recovered from features and deposits dating from the 2nd century CE (Roman period) to the modern period (20th century). The marine mollusc assemblage was dominated by the European flat oyster (Ostrea edulis), with mussels (Mytilus sp.) well represented, albeit by small fragments. Common cockle (Cerastoderma edule) and common limpet (Patella vulgata) were present in small numbers (Table 44).

Taphonomic assessment

Surface preservation varied between 'good' and 'poor' (categories 2–4 on a scale of 1 to 5), with most surviving in 'poor' condition (65.4% by count, n=403). Burnt shell included four fragments of oyster and one fragment of cockle, all recovered from mid-Roman period deposits. Sixty-nine oyster specimens had 'notched' edge damage consistent with shucking, with an additional 74 identified as having possible evidence for shucking.

Evidence for surface modification caused by a range of marine invertebrates was recorded on oyster shells in 244 instances, and included sponge (cf. Cliona celata), annelid worm (Pomatoceros triqueter), spionid worm (Polydora ciliata), bryozoa, and/or barnacle (Cirripedia). Oysters with infestations comprised 35% of the assemblage; in some cases, single specimens were infested by up to four species (Table 45).

Measurements

Many oyster specimens had flaked or fragmented edges that precluded measurement for size reconstruction analysis, beyond estimation of size using a diameter chart (Savine 2024, 523, Appendix 7). In total, 924 oyster valves were suitable for size estimation by the diameter chart method (Table 46). The valves represented specimens ranging in size from very small to very large, with small and medium specimens forming 75.5% of the measurable oyster valves (n=194, minimum number of valves (MNV)).

Quantification

While fragment count (Table 46) provides an overview of the marine mollusc remains at the site, variation in fragmentation between the species recorded means that the data is not representative of the frequency of the different taxa, and therefore may not be indicative of the importance of different shellfish resources at the site. Site-wide quantification using minimum number of valves (MNV) and minimum number of individuals (MNI) for bivalves based on umbo presence, and minimum number of individuals for gastropods based on apex presence, is presented in Table 47. Quantification by minimum number of individuals suggests that the oyster was the most frequently consumed shellfish species at the site, while mussel, cockle, and limpet formed rare additions to the diet (Table 47). These data will be considered below by phase of occupation at the site.

Results

The marine mollusc remains derive from human activity at the site from the Roman to modern periods, 2nd century CE (Phase %) to 20th century (Phase 41). These are presented by chronological period, below. Marine mollusc remains from modern (Phase 41) and unphased deposits are not discussed further.

Roman period (Phases 5 to 16)

Marine mollusc remains were recovered from Roman deposits and features dating to the 2nd and 3rd centuries CE (Table 44). The Roman period assemblage was dominated by mussel fragments (n=155) and edible oyster (n=118), though as discussed above (Table 47) the fragmentary nature of the mussel remains has exaggerated the level of their consumption at the site. Four fragments of edible cockle and a single limpet were also recovered from Roman period contexts, Table 44.

Medieval period (Phases 18 to 34)

Medieval period deposits and features also contained abundant marine mollusc remains, though the focus at this time was on the edible oyster (n=237), which formed 94.8% of the medieval period marine mollusc remains by count (Table 44). Twelve fragments of mussel shell and one limpet were the only other marine shellfish represented in medieval phases of activity.

Post-medieval period (Phases 36 to 39)

Marine mollusc remains from post-medieval phases of activity continued to be dominated by edible oyster (n=40), which formed 88.9% of the post-medieval period marine mollusc assemblage, Table 44. The only other marine shellfish remains from post-medieval contexts were five fragments of edible cockle shell.

Discussion

The terrestrial shell (Table 43) does not contribute to our understanding of human activity or the natural environment at the site, being a very small assemblage comprising ubiquitous species. No further comment is made on these remains.

The marine shell evidence from the Guildhall indicates that the European/edible oyster was the most frequently consumed shellfish at the site from the Roman to modern periods, with peaks in its abundance at the site in the medieval period between the 11th and 14th centuries. The oysters were generally small to medium in size (c 4.6 to 7.5cm in diameter, Table 46), with many having evidence for shucking and infestation by multiple parasites (Table 45). Other shellfish, including mussel, cockle and limpet formed an extremely small part of the overall assemblage, demonstrating that they were likely eaten very rarely at the site.

The European/edible oyster was a valued food resource in Britain in the Roman period (e.g. Cool 2006), and fluctuated in popularity, availability and status from that point onwards to the present day (e.g. Thomas et al. 2020). While the assemblage is of a reasonable size overall, if the period over which the remains accumulated is considered, possibly up to 1000 years, the shells represent only very occasional consumption of oyster, with this occurring consistently during the periods of activity at the site. This may be explained in part by the distance from the coast where oysters were harvested to the inland city of York (c 35km), though the River Ouse was navigable from the East Coast/Humber Estuary to York for transportation of goods. However, as discussed by Winder (2015, and references within), the recycling of oyster shells for a wide variety of uses, including for mortar, making lime, poultry grit, shell-tempered pottery, medicines and cosmetics, and their use as fertilizer for calcium-neutralising acid soils, all cause the shells to delaminate and disintegrate or move the shells away from the consumption site. In considering these potential uses for oyster shell it might be expected that shells would be collected in one location in great numbers, or that they were broken down or subjected to processes that would cause their fragmentation, disintegration and eventual loss. While the evidence from Guildhall cannot prove nor disprove that any of these activities took place, it is important to note that the importance of this dietary resource may be underrepresented by the remains recovered.

The detailed study of oysters from Roman and medieval sites has focused on the midlands and southern Britain (see Winder 2015; Thomas et al. 2020). For York, and many sites in northern England, there has yet to be a detailed synthesis of oyster and other shellfish remains for any period, a study which is much needed to further our understanding of human diet and the trade networks with coastal towns.

7.5 Plant Macrofossils and Charcoal by Stacey Adams

Introduction

Following the assessment of 23 bulk environmental samples and 28 sub-samples from boreholes from York Guildhall (Adams 2021) a number were selected for analysis of the charred, mineralised and waterlogged plant macrofossils and wood charcoal. The material derives from Roman and medieval activity from a range of pits and dumps and varying depths of three boreholes. The following discusses the preservation of the charred, mineralised and waterlogged plant macrofossils and wood charcoal and their ability to inform on the diet, arable economy and local environment of the site as well as fuel selection and use.

Methodology

The bulk environmental samples, ranging from 0.2 to 40L in volume, were processed by flotation using a 500µm mesh for the heavy residue and a 250µm mesh for the retention of the flot before being air dried. The flots of samples recognised as waterlogged during the flotation process were retained wet. The flots were sorted under a stereozoom microscope at 7-45x magnifications. Identification of the charred, mineralised and waterlogged plant remains was based on observations of gross morphology and surface cell structure and, where necessary, relevant reference manuals (Cappers et al. 2006; Jacomet 2006) were consulted. Quantification was based on minimum number of individuals and nomenclature follows Stace (1997) for wild plants and Zohary and Hopf (1994) for cereals.

The charcoal fragments were sectioned by hand along three planes (transverse, radial and tangential) according to standardised procedures (Gale and Cutler 2000; Hather 2000). Specimens were viewed under a stereozoom microscope for initial grouping and an incident light microscope at magnifications up to 500x was used to further identify the charcoal. Taxonomic identifications were assigned by comparing suites of anatomical characteristics visible with those documented in reference atlases (Schoch et al. 2004; Hather 2000; Schweingruber 1990). One hundred fragments were identified from each sample based on the minimum number of fragments principle for temperate regions proposed by Asouti and Austin (2005).

Results

Charred Plant Macrofossils: Late 11th to Early 13th Century

The charred plant macrofossils were frequent in the late 11th to early 13th century deposits with good preservation in pit C19095 and moderate in dump C19136. Pit C19095 predominately contained cereal caryopses whilst dump C19136 contained similar quantities of cereals and weed seeds. The cereals caryopses in pit C19095 were largely of hulled barley (Hordeum vulgare) accompanied by small quantities of oat (Avena sp.) and a single rye (Secale cereale) caryopsis whilst dump C19136 was predominantly of oat followed by hulled barley then wheat (Triticum sp.). The late 12th to early 13th century pit fill C19075 contained abundant plant macrofossils mostly of hulled barley caryopses, 7% of which had germinated. Oat accompanied the barley along with one or two wheat and rye caryopses.

Charred hazelnut (Corylus avellana) shell fragments were identified within all the analysed features at the Guildhall along with fruits of common plum (Prunus domestica) and hawthorn (Crataegus monogyna) in dump C19136. A potential cultivar of cabbage/ mustard (Brassica/ Sinapis) was recorded in the dump whilst pit fill C19075 contained a cultivated sweetpea/ pea (Lathyrus/ Pisum) seed.

Weed seeds were occasional within the flots from the analysed Guildhall samples and consisted of wild grasses (Poaceae), including bromes (Bromus sp.) and fescues (Festuca sp.), and legumes (Fabaceae) including vetches (Vicia sp.). Weeds associated with sandy acidic soils were present in the form of sheep's sorrel (Rumex acetosella), corn spurrey (Spergula arvensis) and knotgrass (Polygonum aviculare). A number of plant taxa associated with damp or wet conditions were identified: pale persicaria (Persicaria lapathifolia), curled dock (Rumex crispus) and sedges (Carex spp.), including glacous sedge (Carex flacca). Meadow buttercup (Ranunculus acris) and common spike-rush (Eleocharis palustris) are also indicative of wet conditions and are associated with seasonal flooding or waterlogging of arable land.

Mineralised Plant Macrofossils: 2nd Century CE, Late 2nd to Early 3rd Century CE and Late 11th to Early 13th Century

Mineralised plant macrofossils were identified in a small number of samples from the Guildhall. Mineralisation is the replacement of biological tissue with calcium phosphate and is generally biased towards preserving the soft tissue of fruits and herbs (Carruthers and Smith 2020). Roman dumps C19151 and C19201 both contained mineralised bristle-club rush (Isolepis setacea) seeds. Dump C19151 also contained mineralised flax (Linum usitatissimum) and fruit seeds of elder (Sambucus nigra), apple/ pear (Malus/ Pyrus) and an indeterminate fruit. Mineralised wild taxa of common-spike rush and bristle club-rush (Eleocharis palustris) were also identified within the deposit. The 2nd century CE Roman dump C19201 contained mineralised wild taxa of spike-rushes (Eleocharis sp.), bristle club-rush and dioecious sedge (Carex dioica). Mineralised plant macrofossils were recorded in the medieval deposits and consisted of an indeterminate cereal caryopsis and wild seed in pit fill C19075. Late 11th to early 13th century dump C19136 contained a number of apple/ pear seeds along with elder and black mustard (Brassica nigra) as well as wild taxa of common hemp-nettle (Galeopsis tetrahit), spike-rushes, including common spike-rush, bristle club-rush and a wild grass caryopsis.

Waterlogged Plant Macrofossils: 11th to 13th Century and 16th to 17th Century

Waterlogged plant macrofossils were identified in the samples from boreholes BH2, BH3 and BH4 with preservation ranging from moderate to good. There was little discrepancy between the taxa identified within the varying 11th to 13th century boreholes and the 16th to 17th century borehole sample suggesting similar activity and environments.

Edible fruits were identified in all the boreholes of elder (Sambucus nigra) and bramble (Rubus fruticosas). Strawberry (Fragaria vesca) seeds were present in context C9622 in borehole BH2 and a common plum (Prunus domestica) drupe in context C9634 in borehole BH3. Exotic fig (Ficus carica) was recorded in all fills of borehole BH2 and contexts C9632 and C9631 in borehole BH3. Economic taxa identified at the Guildhall also included edible fruits, nuts and herbs. Hazelnut shell fragments were present in most of the borehole samples, except context C9632 in borehole BH3 and context C9642 in borehole BH4. Turnip (Brassica rapa) seeds were similarly common and only not recorded in contexts C9632, C9628 and C9634 in borehole BH3. Herbs of dill (Anethum graveolens), in context C9633, and onion (Allium sp.), in context C9634 were present in borehole BH3 along with carrot (Daucus carota) in context C9628. The oil/ fibre plant of flax (Linum usitatissimum) was recorded in context C9634of borehole BH3.

Wild plant macrofossils associated with arable/ ruderal ground were the most common amongst the waterlogged assemblage and included lesser stitchwort (Stellaria graminae), corn spurrey (Spergula arvensis), corncockle (Agrostemma githago), black-bindweed (Fallopia convolvulus), violets (Viola sp.), cinquefoils (Potentilla sp.), agrimony (Agrimonia eupatoria), fool's parsley (Aethusa cynapium), red hemp-nettle (Galeopsis angustifolia), common hemp-nettle (Galeopsis tetrahit), bellflower (Campanula sp.), narrow-fruited cornsalad (Valerianella dentata), lesser burdock (Arctium minus), lamb's succory (Arnoseris minima), smooth sow-thistle (Sonchus oleraceus), rough hawk's-beard (Crepis biennis), cornflower (Centaurea cyanus), ploughman's-spineyard (Inula conyzae), stinking mayweed (Anthemis cotula), corn chamomile (Anthemis arvensis) and corn marigold (Chrysanthemum segetum). Wild plants associated with wet or waterlogged environs were present in significant numbers in the borehole BH3 contexts including meadow buttercup (Ranunculus acris), waterblinks (Montia fontana), pale persicaria (Persicaria lapathifolia), knotgrass (Polygonum aviculare), common sorrel (Rumex acetosa), curled dock (Rumex crispus), gypsywort (Lycopus europaeus), common marsh-bedstraw (Galium palustre), water-plantain (Alisma plantago-aquatica) and sedges (Carex spp.), includingdioecious sedge (Carex dioica), hairy sedge (Carex hirta), glaucous sedge (Carex flacca), club sedge (Carex buxbaumii) and distant sedge-type (Carex distans-type). Wet-loving taxa were infrequent within context C9642 in borehole BH4. Woodland taxa were present in all the borehole BH3 samples consisting of small nettle (Urtica urens), red campion (Silene dioica), bittersweet (Solanum dulcamara) and grey sedge (Carex divulsa).

A number of the wild taxa identified within the borehole samples from the Guildhall have useful traits that may have been exploited by the population. The leaves of common mallow (Malva sylvestris), common nettle (Urtica dioica), fat hen (Chenopodium album), common chickweed (Stellaria media), bladder campion (Silene vulgaris) and white dead-nettle (Lamium album) are all edible and can be consumed as a salad leaf or cooked like that of spinach (Mabey 1972, 90–2, 94, 100; Dickson 1987; Svanberg 2012, 320). The seeds of opium poppy (Papaver somniferum) are also edible, and the plant has medicinal properties for pain relief (Zohary and Hopf 1994, 128). Henbane (Hyoscyamus niger) and selfheal (Prunella vulgaris) were also used as medicine in the past, the former as an anaesthetic and the latter as an anti-inflammatory (Monckton 2000, 3; Salisbury 1961, 219). Common chickweed, fat hen, small nettle and black horehound (Ballota nigra) are indicative of nitrophilous soils and the presence of high organic content (Carruthers 1995, 6; Salisbury 1961, 294; Robinson and Griffiths 2008, 64).

Charcoal: 1st to 2nd Century, 2nd to 3rd Century and Late 12th to Early 13th Century

The charcoal from the Guildhall was mostly very well-preserved with the majority of the fragments identifiable to genus or species-level. The charcoal in late 2nd to 3rd century CE dump C19124 was less well-preserved with a number of fragments distorted, thereby making identification difficult. Distortion was present in the form of radial cracks and vitrification as well as general distortion caused by thermal degradation during the charring process. Radial cracks were common within the fragments, appearing as blown-up ray cells causing cracks of missing or exploded tissue and possibly reflecting the burning of fresh wood (Fiorentino and D'Oronzo 2010). Vitrification gives the charcoal a glassy appearance and is often attributed to high temperatures and prolonged burning times (Gale and Cutler 2000; Prior and Alvin 1983), although recent experiments claim that it is not induced by such factors and that the cause is still unknown (McParland et al. 2010). A small number of fragments were affected by post-depositional sediment likely associated with fluctuations in the water table during burial.

Oak (Quercus sp.) was the dominant taxon in all the analysed features, predominately deriving from large branch or trunk wood and was the most affected by radial cracks. Alder (Alnus sp.) and hazel were frequent within the 1st to 3rd century CE charcoal assemblages as well as that of the late 12th to early 13th century, as was wood of the apple sub-family (Maloideae). The apple sub-family includes the taxa of apple, pear, hawthorn, whitebeam and rowan. Poplar/ willow (Populus/ Salix) was present in late 2nd to 3rd century CE build-ups C19151 and C19162 and late 12th to early 13th century pit C19075. Ash (Fraxinus excelsior) and plum-type (Prunus sp.) charcoal was recorded in 1st to 2nd century CE dump C19201 as well as a small number of the 2nd to 3rd century CE features. Birch (Betula sp.) was identified in build-ups C19191 and C19162 whilst field maple was present in dump C19124, build-up C19151 and late 12th to early 13th century pit C19075 and elm (Ulmus sp.) was present in dump C19140. Coniferous wood of Scot's pine (Pinus sylvestris) and yew (Taxus baccata) were recorded in build-up C19176 with Scot's pine also present in dump C19140. The majority of the wood at the Guildhall derived from large branch or trunk with less than 3% of the assemblage belonging to that of small branch or twig wood.

Discussion and Interpretation

Charred and Mineralised Plant Macrofossils

The mineralised plant macrofossils from the 2nd to the early 3rd century CE indicate that flax was potentially cultivated as an oil or fibre crop. Fruits of common plum, apple/ pear, hawthorn and elder likely formed part of the Roman diet in Eboracum with that of common plum and apple/ pear potentially cultivated in orchards as horticulture was well-established in this period (Van der Veen et al. 2007, 204). The remaining mineralised 2nd to early 3rd century CE plant macrofossils were indicative of wet/ waterlogged environs and would have been freely growing on the banks of the River Ouse.

The charred plant macrofossils from the late 11th to early 13th century features indicate that hulled barley was the preferred crop at the site with oat possibly cultivated alongside the cereal as a 'dredge' crop (Banham and Faith 2014, 35). Oat and barley were often cultivated as fodder crops in medieval England (Giorgi 2006, 128; Pelling 2011, 150). The sporadic wheat and rye caryopses were probably contaminants of the main mixed crop. Edible fruits and nuts appear to have formed part of the diet at Guildhall, with elder and hazelnut likely collected from the wild. The apple/pear may have been grown in local orchards whilst herbs and vegetables of cabbage/mustard and black mustard are likely to have been cultivated in gardens. The weed assemblage indicates that the arable fields were potentially seasonally waterlogged with the presence of sedges, meadow buttercup and mineralised common spike-rush (Pelling 2000, 326) with the latter suggesting cultivation on damp meadows (Carruthers 2004, 113), probably along the banks of the River Ouse. Corn spurrey, sheep's sorrel and knotgrass are associated with the cultivation of poor sandy acidic soils (Carruthers 2002, 203; Giorgi 2006, 128) whilst fat hen is indicative of nitrophilous soils and may indicate the addition of organic material to this poor soil in the form of manure (Streeter et al. 2009, 34; Stace 1997, 140).

Waterlogged Plant Macrofossils

The waterlogged assemblage suggests the edible fruits of strawberry and common plum were likely to have been cultivated in gardens or orchards whilst bramble, elder and hazelnut would have been collected from the wild (Greig 1996, 215). Turnip, dill, carrot and onion would have been grown as a garden crop (Harvey 1981, 89; Livarda and Van der Veen 2008) to add flavour and to bulk out the cereal diet. Flax and opium poppy were potentially cultivated for their oil properties with flax also possessing fibre properties and opium poppy medicinal qualities (Zohary and Hopf 2000). Fig would have been imported as an exotic foodstuff as it was widely brought over from the continent in the medieval period (Moffett 1991, 14; Parkinson 1656). Fig was similarly identified at Coppergate (Carrott et al. 1996) and St. Saviourgate (Carrott et al. 1998).

The environment signified by the waterlogged assemblage suggests ruderal grassland on wet or seasonally waterlogged ground with high organic content indicated by common chickweed, fat hen, black horehound and small nettle (Streeter et al. 2009, 34; Stevens 2015, 197; Robinson and Griffiths 2008, 64). A number of the wild plants are edible and may have been consumed alongside the fruits, nuts, herbs and cereals at the Guildhall. The leaves of common mallow, common nettle, fat hen, common chickweed, bladder campion and white hemp-nettle were cooked similar to spinach whilst the seeds of fat hen and opium poppy were often added to bread for texture and flavour (Zohary and Hopf 1994, 128; Stokes and Rowley-Conwy 2002, 97; Mabey 1972, 90–2; Svanberg 2012, 320). Similar wet and shrubby grassland taxa were identified directly across the river at North Street (Adams 2022b), albeit with the absence of economic and edible taxa. This suggests similar environs on either side of the river but with the Guildhall evidencing the local human diet and economic crops.

Charcoal

The charcoal, mostly deriving from Roman contexts, was predominately of oak and would have been harvested for its ability to burn for a long time and its high burning temperatures (Austin 2003, 99). Oak is easily subjected to woodland management techniques of coppicing and pollarding (Rodwell 1991; Polunin and Walters 1985) and would have provided a steady supply of fuel and timber wood at the Guildhall. Coniferous wood of Scot's pine and yew were rare within the assemblage, although they burn well the wood may have been reserved for timber and woodworking (Uzquiano et al. 2015, 230; Gale and Cutler 2000; Ratcliffe 1984, 84). Wood of birch, alder, hazel, ash, plum-type and the apple sub-family, all provide excellent fuelwood (Taylor 1981) and were likely to have been opportunistically collected and added to the oak assemblage as additional firewood. The inclusion of poor fuelwoods of field maple and elm (Austin 2003, 99; Edlin 1949) support the interpretation that additional taxa were randomly collected rather than deliberately harvested. Roundwood, from small branches or twigs, formed less than 3% of the assemblage suggesting that the fuel derived from large branch or trunk wood. Contemporary Roman charcoal of oak, hazel, field maple and ash was identified nearby at Aldwark (Adams 2022a) indicating a similar, yet less varied, fuelwood exploitation across the city. No distinctions can be made across the Roman charcoal assemblage due to the similarity of the deposits across the occupation. The same is true of the late 12th to early 13th century pit C19075, where oak dominated and was accompanied by several other taxa.

The charcoal assemblage indicates the exploitation of a managed yet shrubby oak woodland for fuelwood with the wood of hazel, plum-type and the apple sub-family growing alongside the pollarded or coppiced oak. Elm, field maple, ash and birch are light-loving trees and would have derived from more open landscapes (Pelling 2012; Austin 2003, 101; Rodwell 1991; Polunin and Walters 1985; Taylor 1981, 46) whilst poplar/ willow and alder are endemic to wet environs (Taylor 1981, 45, 54; Pelling 2012) and would have thrived along the banks of the River Ouse and its tributaries.

Conclusion

The Roman occupation at the Guildhall saw the consumption of wild or horticulture fruits of common plum, apple/pear, hawthorn and elder as well as the fibre/oil crop of flax. The environment appeared to be largely waterlogged marshy ground, as indicated by wild taxa. Oak was the predominant fuelwood during the Roman occupation accompanied by taxa from shrubby wooded, open and riverine environs. The exploitation of fuelwood is similar across Eboracum and during the late 12th and early 13th century at the Guildhall.

The medieval occupation of the site relies on a 'dredge' hulled barley and oat crop, likely for use as fodder that would have been cultivated on poor sandy acidic soils potentially improved by manuring. The local environmental was of shrubby ruderal grassland that was waterlogged or seasonally flooded by the River Ouse and this is echoed at North Street across the river. The medieval diet consisted of wild and cultivated fruits, nuts and herbs as well as the collection/ cultivation of potential 'wild' plants for green vegetables and calorific seeds. Flax and opium poppy were potentially cultivated for their oil, fibre or medicinal properties and fig would have been imported from the continent suggesting a relatively high status to the site.

← Previous chapter | Next chapter →