Because many sculptures in public collections were removed from their original sites long ago, scientists and art historians have collaborated to answer questions concerning their geographic origin and attribution. One way in which scientists contribute to solving these problems is by determining the stone’s composition using neutron activation analysis.

Small samples of stone removed from a sculpture, monument or quarry are irradiated in a nuclear reactor to produce radioactive isotopes of elements present in the stone. These isotopes decay by emitting energy which can be measured to identify and quantify the elements in the stone. Stones from different sources may have distinctive compositional patterns – “profiles” - with which data for a sculpture under study may be compared.

This compositional data is added to the Limestone Database which contains samples from sculptures in museum collections, from quarries in the Ile-de-France, Normandy, Burgundy, Périgord, and the Nile Valley, as well as from French monuments and British cathedrals. Compositional information in the database is used to group sculptures and relate them to quarry stone by using multivariate statistical techniques.

Data stored in the Database allow scholars to:

  • group sculptures thought to have a common origin, based on the hypothesis that similar composition implies a common stone source
  • relate sculpture of unknown origin to a monument or to the region or quarry that may have furnished stone for its production;
  • elucidate the compositional differences in stone among quarries from which medieval craftsmen took their raw material;
  • answer questions about the history of construction of monuments made of limestone.

Typical questions that might be answered by compositional analysis:

  • Is a museum piece carved of stone from the same source as that which provided raw material for sculpture in a certain monument?
  • Were all sculptures in this group carved of stone from the same stone source and therefore perhaps for the same church?
  • Does this group of samples from a nave differ from that group of samples from the choir of a certain church? (A change of quarry may indicate a change of building campaign, leading to a refined estimate of the break between building campaigns.)
  • Here are some samples from early work and samples from restorations or later construction. Does this sculpture belong to the first or the second group?

Questions of the type “Where does this come from?” are difficult to answer. The Limestone Database has strengths and weaknesses in geographic distribution. It is best to give us an idea of provenance; then we may at least be able to say “possibly yes” or “pretty definitely no”.

For more discussion, explanation, illustration and examples (including a short video of the entire procedure), see our Examples page and our Bibliography.