The application of microfossils in assessing the provenance of chalk used in the manufacture of Roman mosaics at Silchester

https://doi.org/10.1016/j.jas.2008年03月01日0 Get rights and content

Abstract

Microfossils recovered from chalk tesserae in mosaics from the Roman town of Calleva Atrebatum, modern Silchester, southern England, are used to suggest a provenance for the source-rock. The microfossils include foraminifera characteristic of late Cretaceous (Campanian) foraminiferal biozone BGS20 (quadrata macrofaunal biozone) and foraminiferal subzone BGS21i (basal mucronata macrofaunal biozone). Calcareous nannofossil assemblages from the same tesserae are poorly preserved, preventing precise age determination, but confirm an age of Santonian to Campanian . As indurated chalk beds of this age are not normally present in the stratigraphically higher chalks of southern England, it is probable that calcretised chalk, formed by secondary calcification beneath Palaeogene rock cover, was used to manufacture the tesserae. This suggestion is supported by a comparative analysis of chalk tesserae from the Norden Roman site in Dorset. Although the provenance of the chalk in some of the Silchester tesserae can be placed only within a broad geographical area of downland in southern England, others may have originated in the Dorchester–Swanage area, some 100 km to the southwest of Silchester, or the Portsdown area of southern Hampshire. None of the tesserae have been constructed from chalk found near Silchester.

Introduction

Microfossils are used widely by geologists to determine the relative age of rock successions for the Phanerozoic Eon, the last 542 million years of geological time (see Gradstein et al., 2004), and sometimes for rocks of still earlier periods of geological time (e.g. Sergeev, 2006). This process is based on the identification of species that characterise rock successions formed during successive time intervals. Thus, the Cretaceous chalk rocks that crop out over wide areas of England are characterised by successive (evolutionary) fossil faunas and flora that can be used to correlate rocks of similar age between the Yorkshire coast and southern England. This methodology has been widely applied to the rock succession of Britain since the time of the engineer geologist William Smith in the early 19th century (Winchester, 2001).
Microfossils occur in a wide range of rock types including limestone, sandstones and mudstones. Their minute size (generally <1 mm) makes them ideal for use in biostratigraphy, especially with very small samples. For this reason, microfossils are typically used to establish the biostratigraphy of boreholes drilled for oil exploration where rock material recovered is limited or broken up into small chips. An overview of different microfossils and their applications in dating and correlating rock successions is given in Armstrong and Brasier (2005). Microfossil groups used for these purposes include calcareous (e.g. ostracods, foraminifera, nannofossils), phosphatic (e.g. conodonts), siliceous (e.g. diatoms, radiolarians) and organic-walled forms (e.g. spores, pollen, dinoflagellate cysts).
Microfossils have also been recovered from a variety of archaeological and historical remains including building materials, art and ceramics (e.g. Perch-Nielsen, 1973, Quinn and Day, 2007, Quinn, in press) and indeed from Roman mosaics (Jones, 1989). In this paper we apply techniques of microfossil biostratigraphy to establish the biostratigraphical age of chalk rocks used in the construction of Roman mosaics in southern England. Using this information, it is possible to suggest a likely geological provenance of these rock types from the chalk rock succession. We take, as a pilot study, tesserae sourced from Roman mosaics in the town of Calleva Atrebatum, modern Silchester. For comparison, we have also examined chalk artefacts from the Norden Roman site near Corfe Castle, Isle of Purbeck, Dorset.

Section snippets

Mosaics at Calleva Atrebatum

The Iron Age and Roman settlement of Calleva Atrebatum (Silchester) is situated between Basingstoke and Reading in northern Hampshire, southern England (Fig. 1). The town stands relatively high on a gravel-topped spur and looks east and south over the clay lowland. The position of the site was determined to some extent by the presence of gravel, which was a source of water and was easily cleared for agricultural purposes, and contrasts with the heavily wooded clays below the site.

Microfossil groups in the Chalk tesserae

Five loose, chalk tesserae from the ongoing excavations of insula ix at Calleva Atrebatum have been examined for calcareous microfossils (see Table 1). Microfossils most commonly found in the chalk are calcareous nannofossils and foraminifera, but ostracods are also present. The microfossils from the Silchester tesserae are housed in the biostratigraphical collections of the British Geological Survey, Nottingham (registration numbers MPA54160 to 54164, see Table 1), and those illustrated are

Methods for recovering microfossils

In order to release microfossils from the chalk, the tesserae were crushed to about 3–5 mm grains, soaked in white spirit for 30 min and then boiled in sodium hexametaphosphate ([NaPO3]6) to disaggregate the rock. Despite their hard (indurated) nature and incomplete disaggregation, the tesserae yielded sufficient foraminifera and ostracods to permit biostratigraphical age determinations.
For study of the coccoliths ca. 2 mm3 of chalk was crushed, suspended in buffered distilled water, allowed to

Geological setting of Calleva Atrebatum

The town of Calleva Atrebatum stands on the Silchester Gravel, one of a series of Quaternary river terraces associated with the River Kennet (Fig. 1). The solid geology around the town is dominated by Palaeogene rocks, particularly the Eocene (about 34–56 million years old) London Clay and Bagshot Sands formations. The London Clay Formation is up to about 90 m thick in the Reading District and comprises dark grey clays and silty clays with a more sandy uppermost part. The overlying Bagshot Sands

Foraminifera

Foraminifera from the Upper Cretaceous Chalk Group of southern England have been examined in detail (cf. Bailey et al., 1983, Bailey et al., 1984, Hart et al., 1989; and references therein) so that the distribution of species is well known. Wilkinson (2000) adapted and emended existing biostratigraphical schemes and applied them to high-resolution mapping carried out by the British Geological Survey. Foraminifera recovered from the Silchester tesserae were compared with this large dataset.
The

Biostratigraphical provenance of the chalk tesserae from Silchester

Preservation of microfossils varied between the tesserae with the best results for all groups coming from MPA54160 and MPA54161 (Table 1). Calcareous nannofossils indicated a Santonian–Campanian age for these tesserae, the benthic foraminifera indicate foraminiferal biozone BGS20 and subzone BGS21i, both of which are indicative of the Campanian (Fig. 3).
Foraminiferal biozone BGS20 occurs in the uppermost Newhaven, Culver and basal Portsdown Chalk, which can be seen extensively in the South

Comparative analysis of chalk tesserae from Norden

Samples of the source rock from which the chalk tesserae at Silchester were manufactured are unavailable. However, the manufacture of chalk tesserae is attested at the Roman settlement at Norden, near Corfe Castle, Dorset (Fig. 2), where over 1500 examples and associated waste chippings were found on a working floor of the late first century AD. In addition, sawn slabs of chalk for use in wall decoration or as floor tiles were also found from contexts ranging from the second through to the late

Geographical provenance of the chalk tesserae at Silchester

By comparison with the Norden chalk material, the source rock for the chalk tesserae at Silchester was also probably indurated as a consequence of secondary cementation. Calcretised chalk is seen to occur most commonly where the overlying Reading Formation (Palaeogene age) is a mudstone. This calcretised chalk occurs as masses, some of considerable size: blocks up to 2 ×ばつ 1 ×ばつ 0.5 m have been brought to the surface during modern ploughing. Calcrete is hard enough to be cut and polished, has a

Conclusions

The mosaic artisans at Calleva Atrebatum used chalk from the upper part of the White Chalk Subgroup and specifically from foraminiferal biozone BGS20 (quadrata macrofaunal biozone) and foraminiferal subzone BGS21i (basal mucronata macrofaunal biozone). They favoured the hard, marble-like calcrete, which formed by secondary calcification at the top of the chalk, because primary indurated beds of this age are not present in Southern England. They collected the calcrete where the top of the chalk

Acknowledgements

We are very grateful for the constructive reviews of two anonymous reviewers. The authors also thank P.M. Hopson (British Geological Survey), for his advice during the preparation of this paper. We thank Peter Woodward, Curator, Dorset County Museum, Dorchester, for supplying us with material, and Mike Tabecki (British Antarctic Survey) and Jim Riding (British Geological Survey) for undertaking micropalaeontological preparations. I.P.W. publishes with the permission of the Executive Director of

References (38)

  • J.R.L. Allen et al.

    Early Roman mosaic materials in southern Britain, with particular reference to Silchester (Calleva Atrebatum)

    Britannia

    (2004)
  • J.R.L. Allen et al.

    Burnt Kimmeridgian shale at early Roman Silchester, south-east England, and the Roman Poole-Purbeck complex-agglomerated geomaterials industry

    Oxford Journal of Archaeology

    (2007)
  • H.A. Armstrong et al.

    Microfossils

    (2005)
  • H.W. Bailey et al.

    The Coniacian–Maastrichtian stages of the United Kingdom, with particular reference to southern England

    Newsletters in Stratigraphy

    (1983)
  • H.W. Bailey et al.

    Biostratigraphical criteria for the recognition of the Coniacian to Maastrichtian stage boundaries in the Chalk of north-west Europe, with particular reference to southern England

    Bulletin of the Geological Society of Denmark

    (1984)
  • R.H. Bate et al.

    A Stratigraphical Index of British Ostracoda

    (1978)
  • G.C. Boon

    Silchester: The Roman Town of Calleva

    (1974)
  • P.R. Bown

    Calcareous Nannofossil Biostratigraphy

  • P.R. Bown et al.

    Techniques

  • British Geological Survey

    Basingstoke. England and Wales Sheet 284. Solid and Drift Geology. 1:50,000

    (1980)
  • British Geological Survey

    Reading. England and Wales Sheet 268. Solid and Drift Geology. 1:50,000

    (2000)
  • F. Brotzen

    Foraminiferen aus dem Schwedischen untersten Senon von Eriksdal in Schonen

    Årsbok Sveriges Geologiska Undersökning

    (1936)
  • J.A. Burnett

    Upper Cretaceous

  • J.A. Crux

    Upper Cretaceous (Cenomanian to Campanian) calcareous nannofossils

  • B. Cunliffe

    Excavations at Fishbourne 1961–1969

    (1971)
  • E. Erba

    The first 150 million years history of calcareous nannoplankton: biosphere–geosphere interactions

    Palaeogeography, Palaeoclimatology, Palaeoecology

    (2005)
  • M.B. Hart et al.

    Cretaceous

  • Hopson, P.M., in press. Review and classification of the Upper Cretaceous Chalk of England and Scotland, with a...
  • F.M. Gradstein et al.

    A geological time scale 2004

    (2004)

    • Cited by (22)

    • Marine and marginal marine Ostracoda as proxies in geoarchaeology

      2022, Marine Micropaleontology
      Citation Excerpt :

      Sources of ostracods in this context can be contemporary sediments like ship ballast (Ansorge et al., 2011), or fossil sediments like limestones, marls and siliciclastics. Ship ballast is presented in a case study below (paragraph 3.3), other examples are building stones (Wilkinson et al., 2008) (Fig. 1i) or pottery (e.g., Capelli and Bonifay, 2014) (Fig. 1j) containing fossil ostracod associations. These associations allow tracing the origin of the source rocks with sometimes remarkable detail.

    • Ostracodological studies in archaeological settings: A review

      2015, Journal of Archaeological Science
      Citation Excerpt :

      The question of provenance of building materials has also been analysed from a palaeontological perspective, using foraminifera and ostracods in particular. The main aim in this case was to understand how man exploited the surrounding territory for his own purposes (Wilkinson et al., 2008). Lately, both methodologies have been at the same site, to assess the provenance of late Iron Age potsherds and of the clay from the till at the Burrough Hill fort, England (Williams et al., this issue).

    • Provenance of chalk tesserae from Brading Roman Villa, Isle of Wight, UK

      2011, Proceedings of the Geologists Association
      Citation Excerpt :

      Since then the field has advanced considerably. Studies of stone materials dating to the Roman period, including tesserae, have been carried out using microfacies classification (Flügel and Flügel, 1997; Flügel, 1999, 2004), petrology (Allen and Fulford, 2004; Allen et al., 2007), an integrated approach including geochemistry (Hayward, 2009), and microfossil analysis (Wilkinson et al., 2008). The results have been informative.

    • Applications of palaeontology: Techniques and case studies

      2011, Applications of Palaeontology Techniques and Case Studies
    View all citing articles on Scopus
    View full text
    Copyright © 2008 Elsevier Ltd. All rights reserved.