4.2 Coarseware pottery

Leary: Report on Romano-British pottery (Digital Archive)

A total of 1045 sherds of Romano-British pottery (just under 25kg and 18.48 EVES) were identified from the well (see archive). These included small numbers in fabrics of East Yorkshire grey ware from kiln groups such as those on Holme-on-Spalding Moor and Norton-type wares (Halkon and Millett 1999; Hayes and Whitley 1950, or perhaps made nearer York at Stamford Bridge), BB1, colour-coated, shell-tempered, white-slipped and oxidised wares. Forms for these components included one lid, fragments from four beakers, three dishes, seven bowls, two flagons (one face-neck) and a bead-rim deep bowl, plus one lid-seated wide-mouthed jar in calcite-gritted ware (Monaghan 1997: types BF5, DD1, DF3, DF7, DP9, FT, KF2, KM fabric B12 pentice-moulded type, LD, YH5 and JH3 wide-mouthed) and a deep subconical bowl (as Perrin 1981, nos. 386-7, probably from South Yorkshire). These minor types came predominantly from the latest fills above deposit 1979.

The remaining vessels present in the well, the overwhelming majority of material and concentrated in lower fills, comprised jars of three types: calcite-gritted jars with lid-seated rims of Huntcliff type (c. 38% by EVES: henceforth Huntcliff-type jars); grey ware jars with constricted mouths and often with lug handles on the shoulders (c. 28%: henceforth grey ware jars but predominantly Crambeck grey ware); and small, handmade jars with everted or hooked rims, often with acute lattice burnish decoration on the girth, of so-called 'signal station' type (c. 27%: henceforth small, handmade jars) (Figure 7). Nine of the grey ware jars, plus one of the Huntcliff type and two of the small, handmade ones, had a distinct wear pattern around the edge of the base, suggesting they were often placed on rough ground or on a stone. Some were in fabrics that were similar to the East Yorkshire grey ware industries, such as those on Holme-on-Spalding Moor, and thus were either fairly old vessels when deposited or stand as evidence that such industries continued producing pottery on a small scale very late in the Roman period.

Figure 7: Proportions of vessel types (table and graph). Based on percentage of Estimated Vessel Equivalents.
Vessel type Rel % EVES Figure 7
bowl 2.60
beaker 1.19
dish 0.91
flagon 0.81
jar 37.71
small jar 26.73
lid 0.54
narrow-m jar 28.35
wide-m deep bowl 0.54
wide-m jar 0.59

Two complete jars were included in the well assemblage (in fills 2093 and 2046) and are discussed in more detail below. Even without these vessels, the profile for wares and vessel types from the feature are quite different from contemporary groups elsewhere, most obviously in the paucity of tableware, mortaria, amphorae, flagons and beakers. This assemblage also differs in terms of wear patterns and fragmentation, there being more fresh sherds here, a greater number of sherds from each vessel than normal, and unusual wear patterns on base edge and girth of jars. When looked at stratigraphically, the forms in the lowest four fills are similar. The quantity of pottery from each stratum is quite different, however, notwithstanding their different volumes, with a huge concentration of pottery in the two earliest levels. These individual groups are considered in detail next.

Early fills 2109 and 2093 generated assemblages that evidenced a number of cross-joins, implying that they were deposited together. The bulk of material from the latter deposit comprised our three main jar types (see above). These had been used but the edges were not very abraded, and they were broken into many sherds of a size that would not be typical if near-complete vessels were thrown in the well. This level of fragmentation suggests that this material was not complete when deposited (Table 2).

Most of this assemblage therefore contrasts significantly with the most prominent aspect of the ceramic group, a virtually complete Huntcliff-type jar (Figure 8) deposited on the south side of the well, just above the complete bucket. This patterning could be explained if containers directly associated with the use of the well were thrown in it at the end of its life, alongside rubbish from the general vicinity comprising fragments of previously broken jars of various forms (see further discussion below of sooting on Huntcliff-type jars, however).

Figure 8

Figure 8: Complete pot in deposit 2093

Overlying fills 2046 and 1979, in contrast, contained far less pottery than 2109 and 2093 (2046: eight sherds, most Crambeck fabric; 1979: two sherds, one from a Huntcliff-type jar). The former did, however, include a rim fragment from a long-necked beaker, a vessel of a type that is uncommon in late 4th-century assemblages, which are usually dominated by jars. This beaker and the general paucity of ceramics thus distinguish this pair of deposits from those below. In addition, 2046 contained a complete grey ware jar, without lugs, probably belonging to Monaghan's type JW group (Monaghan 1997) and perhaps rather old when deposited. Given its completeness, character and context, this vessel might have some claim to be part of structured deposition practices.

Mainman: Assessment report on possible Anglian material (Digital Archive)

Didsbury: Assessment report on handmade pottery (Digital Archive)

Later fills, taken as a group, provided much smaller assemblages but evidenced a far wider range of material. This included vessels in forms/fabrics not present lower down such as beakers, bowls and dishes. Some of these dated to the 3rd/4th centuries and, in one case, to the 2nd century. The more residual component may have been derived from material through which the feature cut, falling in as the well lining collapsed with other, more contemporaneous components derived from activity near the well head or during its construction. Taken in the round, none of this material distinguishes itself markedly from the pottery circulating elsewhere on the site in the late Roman period. Finally, uppermost deposit 1106 also contained several handmade sherds of possible Anglian date (and several fragments of suggested Iron Age date, whose dating thus probably needs revisiting) (see archive) This post-Roman material implies that this deposit accumulated in what had become a depression above the former well.


Internet Archaeology is an open access journal. Except where otherwise noted, content from this work may be used under the terms of the Creative Commons Attribution 3.0 (CC BY) Unported licence, which permits unrestricted use, distribution, and reproduction in any medium, provided that attribution to the author(s), the title of the work, the Internet Archaeology journal and the relevant URL/DOI are given.

University of York legal statements

File last updated: Fri Jun 21 2013