Constraining the provenance of the Stonehenge ‘Altar Stone’Evidence from automated mineralogy and U–Pb zircon age dating

Authors Organisations
  • Richard Bevins(Author)
    Amgueddfa Cymru | National Museum Wales
  • Duncan Pirrie(Author)
    Prifysgol De Cymru | University of South Wales
  • Rob A. Ixer(Author)
    University College London
  • Hugh O'Brien(Author)
    Geological Survey of Finland (GTK)
  • Mike Parker Pearson(Author)
    University College London
  • Matthew R. Power(Author)
    Vidence Inc.
  • Robin K. Shail(Author)
    University of Exeter
Type Article
Original languageEnglish
Article number105188
JournalJournal of Archaeological Science
Early online date27 Jun 2020
Publication statusPublished - 01 Aug 2020
Permanent link
Show download statistics
View graph of relations
Citation formats


The Altar Stone at Stonehenge is a greenish sandstone thought to be of Late Silurian-Devonian (‘Old Red Sandstone’) age. It is classed as one of the bluestone lithologies which are considered to be exotic to the Salisbury Plain environ, most of which are derived from the Mynydd Preseli, in west Wales. However, no Old Red Sandstone rocks crop out in the Preseli; instead a source in the Lower Old Red Sandstone Cosheston Subgroup at Mill Bay to the south of the Preseli, has been proposed. More recently, on the basis of detailed petrography, a source for the Altar Stone much further to the east, towards the Wales-England border, has been suggested. Quantitative analyses presented here compare mineralogical data from proposed Stonehenge Altar Stone debris with samples from Milford Haven at Mill Bay, as well as with a second sandstone type found at Stonehenge which is Lower Palaeozoic in age. The Altar Stone samples have contrasting modal mineralogies to the other two sandstone types, especially in relation to the percentages of its calcite, kaolinite and barite cements. Further differences between the Altar Stone sandstone and the Cosheston Subgroup sandstone are seen when their contained zircons are compared, showing differing morphologies and U-Pb age dates having contrasting populations. These data confirm that Mill Bay is not the source of the Altar Stone with the abundance of kaolinite in the Altar Stone sample suggesting a source further east, towards the Wales-England border. The disassociation of the Altar Stone and Milford Haven undermines the hypothesis that the bluestones, including the Altar Stone, were transported from west Wales by sea up the Bristol Channel and adds further credence to a totally land-based route, possibly along a natural routeway leading from west Wales to the Severn estuary and beyond. This route may well have been significant in prehistory, raising the possibility that the Altar Stone was added en route to the assemblage of Preseli bluestones taken to Stonehenge around or shortly before 3000 BC. Recent strontium isotope analysis of human and animal bones from Stonehenge, dating to the beginning of its first construction stage around 3000 BC, are consistent with the suggestion of connectivity between this western region of Britain and Salisbury Plain.This study appears to be the first application of quantitative automated mineralogy in the provenancing of archaeological lithic material and highlights the potential value of automated mineralogy in archaeological provenancing investigations, especially when combined with complementary techniques, in the present case zircon age dating.


  • Altar stone, Automated mineralogy, Bluestones, Provenancing, Sandstone, Stonehenge, U–Pb age dating