SGRP home page

The National Roman Fabric Reference Collection: a Handbook

Hand specimen picture panel
Thin section picture panel

References

Appendix 1: Keywords and Definitions
Appendix 2: Physical Layout of Sherds Housed in the NRFRC

 

Savernake Grog-tempered ware (SAV GT)

Seven samples

General appearance

Swan (1975, 42) summarises the fabric as ‘. . . a hard, fine, light blue-grey to grey-buff paste, which often appears darker on the surface, particularly when it has been smoothed or burnished . . . distinctive coarsely-speckled appearance (often lumpy on the surface)’ due to the darker constituents of the tempering. Our samples conform to this, as pale grey (8/0) or grey-buff (10YR 8/1), although two have orange (2.5YR 5/8) cores. The feel is generally rough, sometimes powdery. Unlike the other grog-tempered wares in our collection, most vessels are wheelmade. A range of forms was produced: bead-rim bowls or jars, necked-cordon storage jars and bowls, lids, flagons, flat-rim bowls and dishes copying BB1, and Gallo-Belgic derived platters.

Hand specimen

This is a highly variable fabric, depending on vessel size (Swan 1975, 42). All our samples are grog tempered, with distinctive ill-sorted common or abundant angular grains of grey (frequently pale) and brown or black grog, to 4.5mm. Quartz is also common, varying considerably between the samples, it normally measures between 0.1–0.2mm but is present to 0.5mm. Red iron-rich inclusions, to 1.7mm, can also be seen in some samples. Swan (ibid) describes the fabric as containing white or grey pulverised flint, and although in the hand it could not be distinguished from quartz, flint was visible in thin section. One sample is much finer with only sparse grog to 1.0mm.

Thin section

An abundantly silty matrix with common ill-sorted quartz, usually to c 0.3mm but up to 0.6mm, set in a part-isotropic matrix occurs. The section is distinguished by abundant ill-sorted inclusions measuring between 0.1–1.0mm, and including grog, clay pellets and opaques.

Source

Some actual kilns are known near Mildenhall in the Savernake Forest, but other manufacturing sites undoubtedly existed (Rigby 1982, 154). Swan (1975) has made a convincing case for production at Oare.

Donor

The Museum, Devizes

Museums

Corinium Museum, Cirencester; The Museum, Devizes

References

Annable, F K, 1962 A Romano-British pottery in Savernake Forest, kilns 1–2, Wiltshire Archaeol Mag 58, 143–55

Clifford, E M, 1961 Bagendon: a Belgic oppidum

Hodder, I, 1974 The distribution of Savernake ware, Wilts Archaeol Natur Hist Mag 69, 67–84

Luckett, L, 1970 The Savernake kilns, Wiltshire Archaeol Natur Hist Mag 65, 200–1

Rigby, V, 1982 The coarse pottery, in Early Roman occupation at Cirencester (J Wacher & A McWhirr), Cirencester Excavations 1, 153–209

Swan, V G, 1975 Oare reconsidered and the origins of Savernake ware in Wiltshire, Britannia 6, 36–61

See the related record on the Atlas of Roman Pottery on the Potsherd website

Plate 159: Fresh sherd break of SAV GT (width of field 24 mm). Click to see a larger version

Plate 159: Fresh sherd break of SAV GT (width of field 24 mm)

Plate 159.1: Photomicrograph of SAV GT (XPL) (width of field 1.74 mm). Click to see a larger version

Plate 159.1: Photomicrograph of SAV GT (XPL) (width of field 1.74 mm)

Plate 159.2: Photomicrograph of SAV GT (PPL) (width of field 1.74 mm). Click to see a larger version

Plate 159.2: Photomicrograph of SAV GT (PPL) (width of field 1.74 mm)


<< previous county: West Midlands  ::  South-West White-slipped ware >>