In recent years, studies of historical trends in the use of color have focused on the increasing popularity of blue in Europe during the twelfth century. Notably, Michel Pastoureau, in Bleu, histoire d'une couleur, demonstrates that blue appeared more frequently on stained glass, sculpture, paintings, and illuminated manuscripts, but also in secular contexts, on royal emblems, banners, and ordinary textiles and clothing. The material aspects of blue paint, particularly on sculpture, have also become the subject of investigation. The valuable mineral lapis lazuli, known as natural ultramarine when used as a pigment, has been identified in conjunction with recent restorations of early Gothic facades on the cathedrals of Senlis, Angers, Bourges and Poitiers, and its presence on twelfth-century wooden sculpture from France, Italy, Spain, Germany, and Scandinavia has been confirmed as well.

Lapis lazuli is a semi-precious stone consisting of the blue mineral lazurite, with secondary silicates and inclusions of calcite and pyrite. Exported in medieval times from Badakshan in modern day Afghanistan, lapis was probably routed through the port of Venice. It became known in Europe as azurrum ultramarinum, Latin for "blue from across the sea," and its value is described in contemporary contracts as rivaling that of gold.

Found as veins in limestone or granite formations, lapis lazuli must be extracted, crushed, and purified before it can be used as a pigment. Unrefined ground lapis is a pale gray-blue mixture of lazurite crystals and colorless particles. To obtain a more intense color, medieval tracts recommend two general methods. The first consists simply of washing the ground lapis with water, or a mixture of water and substances such as honey, gum, or vinegar. The second, which produces a superior quality of ultramarine, calls for the addition of the powdered mineral to a wax-resin-oil mixture, wrapping it in cloth, and kneading the mass while keeping it submerged in dilute lye, an aqueous solution of potassium or sodium hydroxide. The impurities remain in the wax mass while the blue particles of lazurite are washed out and settle at the bottom of the vessel. Although the most famous description appears in the fifteenth century, in Cennini's treatise Il Libero dell'Arte, first mention of this method in Europe is found in the Liber Claritas. Probably translated into Latin in the thirteenth century, this treatise is believed to be the work of an Arab alchemist named Jabir, or Geber, who reportedly was active in eighth-century Baghdad.

The Cloisters is fortunate to have in its collection three twelfth-century French wooden sculptures on which the use of ultramarine in the original decorative scheme has been confirmed: the Enthroned Virgin and Child said to be from Montvianex (Figure 1), the Enthroned Virgin and Child from Autun, and the Torso of Christ from a Deposition Group said to be from Lavaudieu (Figure 2). The research presented here focuses on the purity of the ultramarine pigment used on these three works, as well as methods of application. To this end, analysis using energy-dispersive X-ray spectroscopy, Raman microspectroscopy, and polarized light microscopy on both dispersed pigments and cross-sections have been undertaken. The composition of binding media on two of the sculptures was considered as well. The Montvianex Virgin's paint layers were investigated with the use of Fourier transform infrared spectroscopy, high performance liquid chromatography, and gas chromatography/mass spectrometry (GCMS), while cross-sections of the Lavaudieu Torso were examined only with GCMS.

Samples of natural ultramarine removed from the two Virgin figures and the Torso of Christ are all of a purity that suggests at least partial refinement of the lapis. The traces of deep blue found on the Autun Virgin's mantle contain a significant amount of large lazurite crystals that measure up to twenty-five microns, while the remainder range from five to ten microns (Figure 3). Visible in the cross-section are two generous layers of blue paint applied sequentially that together measure in excess of 150 microns. The sample, however, may not be representative since it was removed from the bottom of a fold, where paint was likely to accumulate. The blue layer lies directly on a lead white ground, and the pigment-to-medium ratio suggests a matte, opaque appearance. No media analysis was carried out on the sample because the paint layer was contaminated during a past consolidation treatment, but given the twelfth-century date of the Lavaudieu Torso the binder is most likely an egg or glue tempera.

The binder of the blue pigment present on the mantle of the Montvianex Virgin was examined instrumentally and indeed found to contain egg protein. The ratio of lazurite to gray impurities is less than that observed on the Autun Virgin, although still greater than in unrefined lapis specimens (Figure 4). In addition, the lazurite crystals are smaller, measuring ten microns or less, and the thickness of the paint layer does not exceed forty microns. To increase the intensity of the blue color, and perhaps to allow for a more parsimonious application of the valuable pigment, the paint was applied over a gray preparation layer containing lead white and charcoal black.

The original blue color on the loincloth of the Torso of Christ is now obscured by red overpaint (Figure 5). With the use of GCMS the medium of the blue layer was analyzed, and based on the presence of characteristic fatty and amino acids was identified as egg yolk tempera. The particle size of the lazurite is ten microns or less, while the percentage of gray impurities lies between the values observed on the two Virgin figures. The first layer of natural ultramarine paint, measuring from forty to eighty microns in thickness, was applied over a pale-blue preparation layer consisting of lead white and small, dark blue particles that measures about twenty microns in thickness. Using Raman microspectroscopy it was possible to analyze individual blue particles in this layer, and the results indicate indigo, a dyestuff derived from a variety of plants belonging to the family of Papilionaceae. In twelfth-century France indigo was prepared from either woad (Isatis tinctoria) cultivated in Picardy, Provence, and Languedoc, or from the indigo plant (Indigofera tinctoria), which was grown in India and Persia and exported to France.

Examination of the three twelfth-century sculptures at The Cloisters revealed differences in the purity and particle size of the natural ultramarine, as well as in the preparation layers. The presence of refined natural ultramarine calls attention to the absence of documented techniques for processing lapis lazuli in Europe prior to the late thirteenth century and underscores the need for further research into the origins of purification methods. Although European artists undeniably depended on Central Asia for precious materials such as lapis lazuli, this may not have been the case for the processes used to refine them. To answer this question, not only the occurrence and characteristics of natural ultramarine in European contexts must be considered, but also examples of the pigment on works of art attributed to the many regions along the route from Afghanistan.

In October 2002, Lucretia Kargère was appointed Associate Conservator in the Sherman Fairchild Center for Objects Conservation at The Cloisters, where she has worked as an objects conservator since 2000. She came to the Museum in 1994, when she was awarded the first of several fellowships for the technical study and treatment of medieval sculpture. A graduate of the Institute of Fine Arts, New York University, she received an M.A. in art history and a Certificate in Conservation after completing internships at the Institut Royal du Patrimoine Artistique in Brussels and at private conservation laboratories in Florence and Venice.

Instrumental analyses were carried out by Silvia Centeno, Associate Research Scientist, The Sherman Fairchild Center for Works of Paper and Photograph Conservation and The Sherman Fairchild Paintings Conservation Center, Richard Newman, Research Scientist, Museum of Fine Arts, Boston, and Richard Laursen, Professor of Chemistry, Boston University.