This fragment of gypsum plasterwork (A.79-1922) from the V&A’s collection was found in a house north of the modern city of Samarra, probably by the French archaeologist Henri Viollet (1880-1955), who undertook unofficial excavations at the site in 1910.
 |
| A.79-1922 before conservation (photo: Víctor Hugo López Borges) |
The panel was photographed and documented later by Herzfeld during his Samarra campaigns (1911-1913).
 |
| Herzfeld's photograph, taken on site at Samarra, showing the V&A panel at bottom right (photo from the Freer Gallery website: click here) |
 |
Panels illustrated in Ernst Herzfeld, Der Wandschmuck der Bauten von
Samarra und seine Ornamentik (Berlin, 1923), p.66, fig.90a |
The panel is a fragment of rendered wall, made from gypsum plaster with carved ornamentation. The plaster’s manufacturing technique corresponds to that used in the decoration of many buildings and palaces across the Islamic world, as far west as Spain and Morocco.
Preliminary observations with the naked eye and under a binocular microscope, together with the subsequent conservation treatment, helped to identify the materials present on the fragment and how they were used. It provides a glimpse of how the plasterwork would have looked originally and how the materials were applied and worked.
Description
The fragment is formed of several superimposed layers of gypsum plaster, with a carved front and several renders at the back.
 |
| Layers of plaster in profile view of A.79-1922 (photo: Víctor Hugo López Borges) |
There are up to three layers of coarse gypsum plaster renders at the back, measuring up to 4.5 cm in thickness. These layers would have corresponded to the first work over the wall, when the craftsman applied plaster to render the walls in preparation for the decoration. The layers decreased in thickness as they were applied, so that the thickest is the layer in contact with the wall, followed by two consecutive layers of medium thickness.
The plaster mixture shows large pores as well as voids or air pockets trapped under fresh plaster when it was worked and applied over the wall. The plaster shows numerous large inclusions or impurities, such as quartz, clay, unfired particles of gypsum, and soot. Such impurities and inclusions always indicate a gypsum plaster produced by traditional methods. They can have different origins: mineral impurities naturally occurring in gypsum rocks, the result of the combustion products when the gypsum was fired in the kiln, or contamination during the subsequent grinding and sieving process of the burnt gypsum.
The last layer of coarse plaster applied over the wall was textured to a rough finish, as seen in the top right area of the fragment where the carved plaster is missing. This was done to provide grip for the last layer of fine gypsum plaster, which is the one to be carved.
The design on the front of the panel is carved into a very white and fine gypsum plaster of up to 3 cm in thickness. The plaster mixture is of better quality and of purer gypsum than the type used in the underlying layers. Nevertheless there are still some impurities present, including quartz, unfired gypsum and a small amount of clay. This plaster was applied in two consecutive layers, the top one being the very thin. Once the craftsman reached the thickness desired, the fresh plaster surface was carved with metal tools creating a design in relief. There are marks on the surface that show how these tools cut the plaster and created shapes by dragging the tools along. This probably following an outline previously lightly incised on the fresh plaster surface. The carving goes as deep as reaching the underlying coarse plaster layer. The finer details were simply produced by superficial incisions of triangular section, as we can see on the half moon line carved across the fragment.
Thanks to the cleaning treatment, it was possible to observe on the surface of the panel remnants of a white priming layer.
 |
| Traces of fine white priming layers on surface of A.79-1922 (photo: Víctor Hugo López Borges) |
These were found in deep recesses of the carved decoration where it had managed to survive. The surviving fragments were uniform and thin in section, applied as a wash of fine and pure gypsum probably diluted with water and an organic binding medium. This priming layer would have provided a white and uniform finish to the whole carved wall. It was surprising to observe under high magnification that there were up to 5 superimposed layers of this priming wash, interleaved with reddish particles. This would indicate that the plasterwork was regularly maintained by applying new white layers to refresh the wall and hide the progressive build-up of dirt. This practice of whitewashing gypsum plaster is widely used across the Islamic world even today.
This priming layer would generally have been left unpainted but on occasions could receive a paint layer. It is difficult to establish if such was the case with this fragment, but the reddish particles observed between the priming layers are more likely to indicate dirt deposition on the surface rather than pigment.
Conservation Condition
The fragment was stable although a few fine cracks were observed across the carved plaster layer. These cracks seem to be stable as no movement was evident. Some of the edges are very abraded with loose plaster. The panel has suffered many losses through its history; it has remained in almost the same state since it was documented by Herzfeld, except for a small area of carved plaster at the bottom left corner which is now missing. There are also two areas that seem to have undergone an intervention just after they were found, since they already appear in Herzfeld’s picture (visible on the lower right and middle left sides). They were probably unstable plaster consolidated by applying new plaster and toning it to match the surrounding surface.
The surface was very dirty with heavy superficial and ingrained dirt which obscured the real colour of the plaster. The thicker and more ingrained deposits were red soil resulting from burial, and the clay cemented materials built up by insect activity. The surface was covered with an overall grey dirt layer, from the accumulation of dust and pollution, probably from display in V&A galleries from soon after the fragment reached London in 1921.
The panel also showed numerous small cylindrical holes of different diameters, some associated with the build up of consolidated clay material in the recesses. After careful examination they were identified as the result of termites burrowing through the plaster and building channels along its surface. In some of the carved recesses, the white priming layer was heavily mixed with the deposits of clay and soil particularly from the termite channels.
Conservation Treatment
The surface was cleaned from superficial dirt with the aid of a soft sable brush and a vacuum cleaner with adjustable suction to ensure the surface was not damaged. After this the surface condition of the plaster became more apparent, revealing more areas with abrasions and relatively recent losses where the plaster was powdery and loose.
The areas with powdery plaster were consolidated with Aquazol® 500, a neutral synthetic resin with good penetration properties normally used in the consolidation of gesso grounds. It was diluted in ethanol at 3% in order to minimise the impact on the plaster, as water tends to dissolve gypsum. Several applications with a brush were made until the surface became stable and cohesive.
Some of the thick clay deposits and residues of termite activity were removed mechanically with small tools under magnification. In some recesses more remnants of the original priming layer were found underneath; they were slightly detached from the plaster but still held in situ by the cemented clay deposits. These areas had to be also consolidated with Aquazol® 500 to make them more stable and allow the removal of as much clay as possible. In the end not all the clay could be removed in some areas without running the risk of losing the priming layer.
The surface was then cleaned of ingrained dirt with Anjusil®, a latex poultice containing small amounts of ammonia and surfactants with minimal amount of water. The poultice was applied to the surface with a brush and left to dry.
 |
| Víctor applying the latex poultice (photo: Miriam Orsini) |
 |
| The latex poultice left to dry (photo: Víctor Hugo López Borges) |
This method allows the cleaning of the plaster surface without the use of water, as well as minimising mechanical methods that can be abrasive. The latex releases solvents onto the plaster surface, dissolving pollution and ingrained dirt. Once the latex starts to dry, it reabsorbs the solvents along with the dissolved dirt, which ends up coagulated within the latex. Finally the latex is carefully peeled off.
 |
| View of A.79-1922 during the cleaning process (photo: Víctor Hugo López Borges) |
Several Anjusil® applications where required. In between applications, some of the thickest clay deposits remained slightly wet, becoming softer and more visible, allowing them to be more easily removed mechanically with small tools under magnification. Not all the clay could be completely removed but most of the dirt was lifted off.
The conservation treatment improved the structural and superficial stability of the object, but it also has finally revealed the right colour of the gypsum plaster as well as the remaining traces of the priming layers.
 |
| A.79-1922 after conservation (photo: Víctor Hugo López Borges) |
The treatment as a whole has enhanced our appreciation and understanding of the original characteristics of the materials and techniques present on the fragment. It takes us closer to the craftsman that worked the plaster and to the appearance of the plaster when it originally decorated a wall within the Abbasid palace complex of Samarra.
Víctor Hugo López Borges
Senior Sculpture Conservator, V&A
