4.10 Anglian Pottery Petrological Assessment

by Alan Vince *

4.10.1 Sample selection

Fifty-two samples of Anglian pottery from the West Heslerton excavations were submitted for examination. The samples were chosen at random within a grid defined by the visual fabric classification used as part of the West Heslerton excavation database. Thin-sections were produced of all samples and were stained using Dickson's Method in order to distinguish between ferroan calcite, non-ferroan calcite and dolomite (see also 6.1 The Prehistoric Landscape, 6.5 The Anglo-Saxon Landscape, Is there any evidence for long distance trade and, if so, was this with the 'Anglian' homelands or within a broader North European context? and 7.2.1 Environment and economy).

4.10.2 Objectives

The objectives of this study were as follows:

4.10.3 Pottery fabrics

Fabric Name Heslerton Class Number Sampled
?ANGLIAN unclassified 1
Table 4.14 Fabric classification of the sampled sherds ?ANGLIAN

This sample (L449) contained abundant fragments of coarse-grained sandstone, which was also the main inclusion in two other samples (L424, L425) classified as 'Med Sand'. Coarse-grained sandstone, or the quartz grains derived from such a sandstone, was present in three other samples (L450, L451 and L452) classified as 'MICA & CA/QUARTZITE'. CA + ASHY

The two samples of this fabric group both contained non-ferroan limestone, which could have been a tempering material but contained ferroan microfossils which would have been present in the parent clay. These microfossils were only seen in one other sample, L456, classified as 'CA TEMP'. They seem therefore to form a distinct group, although the two samples differ in the frequency of limestone and in its character. L433 contained fragments of shelly limestone in which the shells were composed of non-ferroan limestone and the matrix of ferroan. L434 on the other hand contained fragments of a non-ferroan limestone breccia. CA TEMP

The eight samples of this fabric group were mostly very similar in their petrological characteristics. All but two samples contained moderate to abundant fragments of sparry ferroan calcite and sparse rounded quartz grains. L455, by contrast, contained moderate angular quartz and brown-stained limestone (ferroan calcite) and L455A contained no calcareous inclusions at all. COARSE SAND

The two samples of COARSE SAND had differing petrological characteristics. L440 contained abundant angular quartz (probably originating as veins in some softer material) whereas L441 contained moderate fragments of rounded quartz, sparse fragments of oolitic limestone, siltstone and sandstone. Both samples, however, contained sparse fragments of basic igneous rock otherwise only found in one sample, L463. FINE SAND

All five samples of 'FINE SAND' contained moderate to abundant fragments of rounded quartz and all but one also contained sparse fragments of sandstone. The clay matrix of all five samples contained sparse fragments of quartz. These similarities suggest that the five samples belong to a single fabric group. MED SAND

These five samples appear to belong to two fabric groups. The first, containing abundant coarse sandstone has been mentioned above (see The second contains abundant rounded quartz. MICA & CA/QUARTZITE

The five samples of 'MICA & CA/QUARTZITE' contain moderate to abundant angular quartz grains, and sparse to moderate fragments of acid igneous origin (almost all felspar). It is possible that these inclusions derive from a felspathic sandstone. MICA/BIOTITE TEMP

The five samples of 'MICA/BIOTITE TEMP' all contain moderate to abundant fragments of acid igneous rock. These were composed of quartz, biotite (some with a distinct bluish tinge), felspars showing perthitic texture and zoning, graphic intergrowth of quartz and felspar. All of these characteristics are found in pottery tempered with Mountsorrel granodiorite from the Charnwood region of Leicestershire. Sample L463, however, also appears to contain fragments of basic igneous rock which are absent from the East Midlands material. MSAX CA GRITTED

The seven samples of 'MSAX CA GRITTED' ware contain fragments of sparry non-ferroan calcite. Two also contain fragments of chalk and in some cases it can be seen that the sparry calcite formed veins within the chalk. Although the main inclusion in this fabric is the same as that in the 'CA TEMP' group, the latter fabric also contains quartz grains which are absent from this fabric. OOLITE TEMP

The four samples of 'OOLITE TEMP' contain moderate to abundant fragments of oolitic limestone composed of non-ferroan calcite. The only other sample that contained oolitic limestone was L441, one of the COARSE SAND samples. STRAW/DUNG

The five samples of STRAW/DUNG all contain organic inclusions probably derived directly from chaff or from dung. Sparse to moderate rounded or angular quartz is present in four samples and sparse fragments of red sandstone in another. Sparse limestone composed of non-ferroan calcite is present in two samples.

4.10.4 Summary

Petrological analysis can, therefore, confirm several of the visually identified fabric groups have a petrological basis. A few of the groups contain one or two 'rogues' which may either represent misclassified sherds or separate fabric groups and a few show such variation from sample to sample as to suggest that they are heterogeneous groups which in reality contain sherds from several sources. To recap, Table 4.15 summarises these conclusions and recommends further analyses.

Fabric Code Major Inclusion Types Comments
?ANGLIAN Coarse Sandstone temper c.2-3 samples require to define group
CA + ASHY Fossiliferous limestone c.3 samples required to define group
CA TEMP Sparry calcite and quartz sand Satisfactory
COARSE SAND Heterogeneous group further analyses and grouping required
FINE SAND Quartz sand Satisfactory
MED SAND Heterogeneous group reclassify all sherds into Coarse Sandstone or rounded quartz groups
MICA & CA/QUARTZITE Felspathic sandstone Satisfactory
MICA/BIOTITE TEMP Acid Igneous rock Satisfactory
MSAX CA GRITTED Sparry Calcite Satisfactory
OOLITE TEMP Oolitic limestone Satisfactory
STRAW/DUNG Heterogeneous group further analyses and grouping required
Table 4.15 Petrological analysis

4.10.5 Provenance studies

As part of this assessment, John Aram visited the Vale of Pickering and sampled various outcrops of oolitic limestone and other rock types with a view to determining the area from which the tempering materials found in the West Heslerton pottery could have been obtained. Further assistance was provided by Dr Ian Freestone of the British Museum Research Laboratory, who both examined many of the thin-sections and gave copies of his unpublished papers on the petrology of Iron Age pottery in the Vale of Pickering, many of which share inclusion types.

From these sources several of the fabric groups found at West Heslerton can be provenanced. For example, the sparry calcite found in two of the groups is ultimately derived from calcite veins in the chalk but is available locally as isolated fragments of calcite. Study of the oolitic limestone samples and comparison with the oolitic limestone found in the pottery will likewise allow this fabric group to be provenanced. The quartzoze inclusions, comprising rounded quartz grains, fragments of coarse-grained sandstone and mixed sands can probably be more closely provenanced through a programme of sand and gravel sampling and the production of test briquettes made by mixing these sediments with a standard, quartz-free clay matrix.

Perhaps the most interesting and intractable problem posed by the petrology of the West Heslerton pottery is that of the acid igneous rock-tempered fabric. Iron Age pottery from several sites in Yorkshire has been shown to contain igneous rock temper. However, in most cases the rocks utilised include basic igneous types derived from the local boulder clay deposits. Not only is basic igneous rock rare in the West Heslerton material but the characteristics of the acid igneous rock are precisely the same as those of the Leicestershire Mountsorrel Granodiorite. Similar fabrics have now been recognised at Wharram Percy and Sancton to the north of the Humber and at Castledyke, Barton-upon-Humber, on the south bank. If a Leicestershire source for the West Heslerton material could be proved it would have profound implications for the degree of mobility and inter-regional contact to be found at that period. On the other hand, if there is a northern source for the West Heslerton material then it has implications for the study of the distribution of pottery which, from its frequency and location, is quite clearly of Leicestershire origin, in that the petrological methods being utilised by David Williams and myself will have been shown to be insufficiently powerful to distinguish the sources. It is proposed, therefore, that a programme comprising Neutron Activation Analysis and chemical analysis of specific minerals be undertaken to provide an independent check on the petrological results.

4.10.6 Technological data

The two areas in which petrology can be used to determine the technological treatment of the pottery are the comparison of potting clays and tempers to determine whether or not inclusions are temper added by human agency or natural components of the clay deposit, and the study of the optical characteristics of the clay matrix to see whether or not the vessel from which the sample came was fired to such a temperature as to produce an optically isotropic ceramic matrix from the original mixture of clay minerals. Addition of tempering

The study of tempering can be carried out by analysis of other fired clay artefacts (such as loom weights and daub), and of clay samples collected in the field as well as by studying the clay matrix of the thin-sectioned samples. These will show whether or not clastic material occurs naturally in the clays and, if so, of what types. In most cases it is already clear that the inclusions over, say, 0.25mm across are of detrital origin but they could as easily be present by natural contamination of clay deposits with material from overlying sand deposits as be the result of deliberate action by the potter. Firing conditions

The clay matrix of all 52 samples was examined and in all cases the matrix was still anisotropic, indicating a short, low-temperature, firing. Matrix classification

The presence/absence of silt-sized quartz and white mica particles was found to vary from fabric group to fabric group. Quartz and/or mica silt is present in CA TEMP, FINE SAND, MICA & CA/QUARTZITE and MICA/BIOTITE TEMP groups only. Ferroan calcite microfossils are present in the CA + ASHY, CA TEMP and MICA & CA/QUARTZITE groups only. This confirms that not only are discrete sources of tempering material used but also at least three separate clay sources.

4.10.7 Conclusions and recommendations

The objectives of the assessment have been achieved with varying degrees of success. As shown in Table 4.15, the petrological analysis has shown that whereas the majority of the petrologically defined groups were recognised by eye, there is a need to study further samples in some cases and re-examine and sample in others. In all likelihood a further batch of 50 samples would probably suffice but it is difficult to make an accurate assessment of the requirements before the heterogeneous groups are re-examined.

The initial work on provenancing the fabric groups indicates that the fabric groups originate in different areas, using at least three basic parent clays and at least eight different sources of tempering material. It is likely that in most cases a source within 20-30km of West Heslerton could be found and, to achieve this, a programme of clay and sand sampling should be initiated covering an area c.30km diameter centred on West Heslerton. Cost and time could be kept to a minimum by sampling, preparation and analysis of the briquettes in bands of c.5km width starting at West Heslerton. Sampling would continue until potential sources have been found for all eight tempering materials and all three clay types or until the 30km band has been reached.

The study of the Acid igneous rock-tempered pottery is a major priority within Anglo-Saxon pottery studies, for the reasons presented above. A programme of Neutron Activation Analysis and Chemical Analysis of specific inclusion types would almost certainly produce a positive result, either definitely proving that the West Heslerton finds were of Leicestershire origin or demonstrating that a second, northern, group of Acid Igneous rock-tempered pottery existed. Should the latter result be obtained then it would become necessary to follow this analysis with further studies to isolate the source.

The proposed work would be of value not only in aiding the understanding of the West Heslerton settlement and the interactions of its inhabitants with those in other areas, but would also form a solid foundation from which to study the entire Anglian settlement of Yorkshire and its relationship with surrounding regions, especially that to the south.


© Internet Archaeology URL:
Last updated: Tue Dec 15 1998