This report contains an assessment of the Roman pottery from the excavation. The report was written by Amy Thorp of the Museum of London Archaeology, but has been edited for web by Chaz Morse and Guy Hunt. For a full index of the assessment reports, please see the assessment report index.

Site archive: Quantification and description

Class Quantity
Stratified Roman pottery 14010 sherds. Weight 260.795kg


All stratified Roman pottery was spot-dated from the site; this comprised 14,010 sherds from 475 contexts. The vast majority (339) of the contexts contain small groups (less than 30 sherds). Of the remaining contexts 118 are medium sized (30-99 sherds), and 18 are large sized (100+ sherds). However, it should be noted that these small groups include some of the most significant material from the assemblage (containing complete or semi-complete vessels from burial and cremation contexts).

The condition of the pottery is very variable depending on context. It is evident that some areas of the site have been heavily disturbed, and some cremation vessels in particular (despite accounting for complete or semi-complete vessels) are highly fragmented. Numerous vessels have significant abrasion of their surfaces; this is primarily a result of the acidic burial environment common to cemeteries.

Dating of the site will need much further examination at analysis, but current indications are that activity begins in the 2nd century AD followed by a late Roman phase (possibly focusing on the 3rd century AD to early 4th century AD, though this is tentative). A further six cremation samples (mostly complete vessels) have yet to be analysed and put into sequence, as they had not undergone osteological examination at the time of this assessment. The vessels are from the following contexts: (961), (1745), (1820), (1837), (1849), and (2009). The jar from context (1837) will be important as it cuts into cremation Subgroup: 73, and is potential dating evidence of the sequence in this area of the site.

A previous excavation phase at Prescot Street in 2006 also recovered 299 sherds, weighing 2.140kg (Richardson 2006). Context (317) contained the most interesting material in the form of 3 complete/semi-complete flagons produced at the Much Hadham kilns of north-east Hertfordshire (MHAD). It is vital that these vessels are re-examined as part of the overall cemetery assemblage at analysis, even if they are not from an actual burial subgroup.


The pottery was spot-dated using standard Museum of London Archaeology methods. It was quantified by sherds, weight and estimated number of vessels (ENV). Percentages presented in the course of this report are based on sherd count unless otherwise stated.


The date ranges from the Prescot Street assemblage clearly show a lack of early Roman activity, with an influx of activity on the site from the early 2nd century AD onwards. However, dating of the cemetery itself should be treated with caution until further analysis of the stratigraphy in combination with individual burial/cremation groups is undertaken. Despite the prevalence of early 2nd-century AD groups, the height of activity at the cemetery is more likely to be later. Residuality of artefacts utilised for accessory goods in burials or cremation urns is a well known practice (Barber and Bowsher 2000: 8). Indeed accessory vessels in inhumations are probably mid-late 2nd century AD onwards, when this practice was more common. The overall dating of the cemetery appears likely to closely follow that of the adjacent site of West Tenter Street; with cremations taking place in the 2nd century AD to mid-3rd century AD and inhumations running right through into the 4th century AD (Whytehead 1986: 122).

An important feature of the assemblage is that only a small proportion of it has been associated with actual burials and cremations (approximately 24%, including ‘possible’ groups). This is not an uncommon feature of cemetery assemblages, and it is probable that some of the pottery is likely to be rubbish dumped on the site from elsewhere in the city. A similarly low proportion of burial furniture was also identified at West Tenter Street (Pierpoint 1986: 68). Disturbance of burials is a further similarity between the two sites, and more burial vessels will be among the general assemblage. Certain vessels already stand out from the assemblage as likely to have been disturbed from burial or cremation groups. The aforementioned flagons from (317) are a prime example.

The range of fabrics and forms present among the Prescot Street assemblage is relatively wide, but with distinct concentrations which are undoubtedly caused by its function as a cemetery. There is strong potential to compare such trends with other sites in the Eastern London cemetery, especially West Tenter Street and St Clare House.

Early Roman

Early Roman material (1st century AD) is primarily fragmentary and abraded, it is clear that pottery from this period is likely to be residual. There are fabrics and forms produced during the early Roman period present, but classic 1st-century AD indicators such as Hoo Island white-slipped ware (HOO) and bead-rimmed jars (2A) are almost absent from the site.

2nd century-3rd century AD

A strong presence of 2nd-century AD material is clearly evident among the Prescot Street assemblage. Black burnished wares (produced from the early 2nd century AD onwards) are a very noticeable feature, and account for 15% of the assemblage. It should be noted that the sherd count of these wares produced a large figure partially down to high fragmentation of many vessels. However, black burnished wares were evidently still a popular choice from the ENV (estimated number of vessels) and their frequent presence in both burial and cremation subgroups.

Possible cremation contexts (to be further analysed alongside osteological information) are interestingly almost exclusively black-burnished wares or Verulamium region white ware (VRW) jars. Prevalent black burnished ware forms support an emphasis on earlier products of these industries. The occurrence of black-burnished-type everted-rimmed jars (2F), produced until around AD 250, is extremely noticeable; only necked jars (2T) occur in equal measure among the variety of jar types. By contrast there is only one example of the later cavetto rim variant of this form in context (1032), burial Subgroup: 11. Similar patterns can be observed in the proportion of bowl types present in black burnished wares.

The clear dominance of fabrics such as Highgate Wood ware C (HWC) in the reduced wares, and Verulamium region white ware (VRW) among the oxidised wares is also likely to be the result of 2nd century AD activity. Products of the latter industry are known to be a preferred choice for cremation vessels (Evans and Pierpoint 1986: 206).

Flagon variants present at Prescot Street are a further interesting feature relating to this period. Late 2nd century AD variants of the ring-necked flagon (1B) are relatively common. Overall Flagons are a strong component of the pottery, despite their relative absence from burial or cremation groups; this may provide further support for ideas of feasting or visits to graves by family taking place at London cemeteries (Bentley and Pritchard 1982 ; Evans and Pierpoint 1986; Hall 1996). The proportions and placement of flagons and other drinking vessels is a feature that should be re-examined at publication.

A reasonable quantity of 3rd-century AD pottery is potentially present among the assemblage. The presence of both central Gaulish and east Gaulish samian wares (SAMCG, SAMEG), including specific forms such as the Ludowici form SMb/SMc bowl (4LUDSM) is notable. Samian ware vessels from these industries have been identified within burial subgroups as well as within general debris. Subgroup: 4, context (740) includes a complete example of Dragendorff form 32 dish (5DR32). Subgroup: 29, context (1735) has seven sherds of a Dragendorff form 38 bowl (4DR38) which account for the whole vessel aside from two small fragments. Samian wares as grave goods are relatively rare, and their presence in inhumations does not appear to occur elsewhere in the Eastern cemetery (Pierpoint 1986; Symonds 2000).

Fine ware Nene valley colour-coated ware (NVCC) beakers are also prevalent, and a good proportion of these are likely to be from the 3rd century AD (with the exception of later pentice beakers). Burial Subgroup: 8 context (869) contains a more unusual albeit fragmented 3rd century AD fine ware, a central Gaulish/Lezoux black colour coated ware beaker with white barbotine decoration (CGBL 3 WBAD). Furthermore, the Much Hadham oxidised ware (MHAD) flagons in context (317) are identical with those in a 3rd century AD grave group at St Clare House, Minories (Harden and Green 1978: 163-175). Sherds of Camulodunum form 306 bowls (4C306), often associated with 3rd-century AD groups, are an unusual presence on Prescot Street. Despite there only being a few sherds of this form in the assemblage, they are important as the bowl is not usually found within the Eastern cemetery (Symonds 2000: 124).

Late 3rd century-4th century AD

The proportion of contexts from the assemblage allocated to this period is relatively low. However, pottery from the late Roman period accounts for some of the most significant groups from Prescot Street. The highlight of the assemblage is the selection of complete and semi-complete Alice Holt Farnham ware (AHFA) flagons, beakers and miniature jar found in cremation Subgroup: 73, (1836). These AHFA vessels were deposited alongside glass vessels, including the rare mosaic glass bowl which is now thought to be unique to the Western Roman Empire. The pottery vessels are evidently a deliberately selected group, being of a similar size and colour (including application of black slip across portions of the vessel); it is very probable that they came from the same kiln firing, a specific order perhaps as all appear unused and in excellent condition. The group consists of three flagons (one with burnished decoration) of varying rim type, three beakers (two of which are similar to pentice beakers (3L)), and finally a miniature jar imitating the later everted-rimmed jar with cavetto rim (2F13). A long rectangular slot has been cut into the shoulder of one of the three flagons (directly opposite the handle). Interpretation of such holes in vessels has been considered a deliberate ritual act ‘killing the vessel’. However, this has been questioned recently and the practice of deliberately holing vessels was probably widespread with other non-ritual purposes. Rectangular slots as seen on this flagon have been explained as money-boxes, and in some cases have been found with coins inside (Fulford and Timby 2001: 296). A tentative date of AD 270–400 has been given to the group so far, but an end date during the early 4th century AD is quite likely. Further research into these vessels will definitely be required at publication. Given the significance of this group it is recommended that they are both drawn individually and photographed as a group (this will allow the rim variations on the flagons to be seen in detail).

The best preserved burial group (Subgroup: 46, (2128)) is also late Roman. The group has three vessels in total; a complete Nene valley colour-coated pentice beaker with rouletted decoration (NVCC 3L ROD), a semi-complete black-burnished-style ware shallow simple dish (BBS 5J), and a complete unsourced sand-tempered ware shallow simple dish (SAND 5J). The latter dish may also be a black burnished ware, but the wear/abrasion on the vessel has made identification difficult. The date of this burial is led by the pentice beaker, which has a range of AD 180–400. However, Perrin (1999: 97) has identified that the form may in fact be restricted to the 4th century AD. It should be noted that in London this is the popularity of this form is from the late 3rd century AD onwards. A further example of a NVCC pentice beaker (also with rouletted decoration) was found in cremation Subgroup: 76, (1822) and (1823); though in this case the vessel was fragmented and only the base had survived.

Analysis of potential

The Prescot Street Roman assemblage has several avenues for further analysis and research. Once the assemblage has been integrated with the full stratigraphic sequence and coins the Roman pottery has good dating potential. However, it has been already highlighted that problems of residuality of burial furniture will need to be considered in the cemetery sequence.

Comparison with other sites in the Eastern London cemetery has strong potential. The selection of types of burial furniture (vessel type and fabric) is important in continuing our knowledge of pottery use within the London cemeteries. Pottery types present in the assemblage outside burial and cremation groups should also be analysed; furthering knowledge of activities taking place associated with the cemetery such as feasting. The adjacent site of West Tenter Street is vital within this analysis, and for examining the dating sequence between the two sites.

Significance of the data

International and National

The selection of the Alice Holt Farnham ware (AHFA) vessels in cremation subgroup 73 is likely to have national significance; primarily this is because of their combination with the rare mosaic glass bowl.

Regional and Local

As part of the Eastern London cemetery the Prescot Street Roman assemblage has local and regional significance. The range of pottery through the assemblage and selection of individual vessels for burial contexts is important in understanding trends within the Eastern cemetery, and comparison with other cemeteries. The preservation of late Roman burial and cremation groups is very significant both for London research and further afield.