Fabrics similar to the one described here have a wide distribution throughout southern Britain, although they could come from a range of source areas with similar natural resources. While form parallels abound in the literature, it has not been possible to cite additional museum collections for the fabric until a firmer source can be defined.
This fabric is variably fired, frequently in shades of dark grey to black (3/0–2.5/0), occasionally with margins either of these same shades or brown (10YR 6/2, 5YR 3/2) to brown-red (5YR 5/4) and with mottled surfaces. The forms and decoration are distinctive, typically hook-rim jars with narrow rilling and flanged bowls. It is a hard fabric with hackly break and smooth to occasionally soapy surfaces.
The fabric is characterised by ill-sorted fossil shell inclusions, normally 0.2–3.0mm, but up to 5.0mm. Other calcareous inclusions are sparse, but some fossiliferous limestone and microfossils can be identified in the hand specimen, as can common matrix-coloured clay pellets (0.5–3.0mm) and sparse quartz (<0.7mm). Although the fabric was previously known as ‘Calcite-gritted’ ware, no calcite is present in the clay.
Both samples were analysed by Mr John Cooper, who identified oyster shell, oyster and brachiopod shell, as well as voids representing burnt-out wood or vegetable remains. The shell inclusions are naturally occurring in the clay, rather than added as temper, and probably derive from Jurassic deposits.
Abundant ill-sorted fossil shell and fossiliferous limestone, and rare microfossils are set in a clean matrix with only rare silt-sized inclusions. Calcareous fragments measure between c 0.1–3.0mm, but normally do not exceed c 1.0mm in size. Other inclusions comprise sparse black opaques, probably equating to Cooper’s organic inclusions, matrixcoloured clay pellets and rare quartz, sometimes polycrystalline.
These samples are not allied with any particular kiln, but Cooper has identified their production area within Jurassic deposits. These beds form a wide belt from Somerset through Northamptonshire to north Yorkshire, but distribution of the ware indicates a source in southern England. Similar types have been tentatively allied with production at Harrold where rilled vessels were also made (Tyers 1996, 192–3). Our samples do not support this, as the Harrold products are consistently pale yellow or oranges, apart from some 1st century sherds described as red-black (2.5YR 2/1, Brown 1994), whereas ours are consistently dark. It is, of course, possible that the more successful of the Harrold products, which travelled away from the kiln site, were reduced (M Lyne, pers comm)
Museum of London
Museum of London
Brown, A E, 1994 A Romano-British shell-gritted pottery and tile manufacturing site at Harrold, Beds, Bedfordshire Archaeol 21, 19–107
Drury, P J, 1976 Braintree: excavations and research, 1971–76, Essex Archaeol Hist 8, 1–43 (3rd ser)
Sanders, J, 1973 Late Roman shell-gritted ware in southern Britain, Unpublished BA thesis, University of London
Tyers, P A, 1996 Roman pottery in Britain
Wallace, C, 1993 Notes on the dating of late shell-tempered wares in Essex, J Roman Pottery Stud 6, 123–6