Technological Observations on the Thera Wall-Paintings

The author concludes that the Thera wall-painting was probably done al secco, although there are indications suggesting the use of a mixed technique in places.

INTRODUCTION

The word "fresco" is nearly always used nowadays in place of the term "wall-painting". Among non-specialists everwhere it has become common practice to regard the two words as synonymous, in spite of the fact that this usage bears no relation to the facts.

The word fresco is Italian, meaning "fresh" or "moist", and it has come by extension to be used of wall-painting carried out on the still fresh surface of the wall. In this kind of painting the fresh lime of the surface coating acts as the bonding agent between the paint and the wall. The fresh lime reacts chemically with the atmospheric carbon dioxide to produce carbonate of lime, thus fixing the pigments in the plaster. This lime paint bond is exceedingly resistant and is capable of lasting for thousands of years.

Parallel with wall-painting on a fresh surface (νωπογραφία, buon fresco)there exists the technique of wall-painting on a dry surface (al secco). Here the role of bonding agent between paint and wall is taken by a variety of organic, animal or vegetable substances.

A combination of the two techniques occurs in the well-known Roman wall-painting. In this we find the painting executed on a fresh white lime undercoat, combined with the use of organic substances.

Many parallels to these two techniques have existed throughout the ages. Nevertheless, buon fresco and al secco painting remain the two basic methods from which all the others have derived. The "encaustic" method, using melted wax as the bonding agent, is mainly used for portable ikons (e.g. Fayum) and only to a lesser degree for walls or rock.

MAKING A WALL-PAINTING

All the books we read on wall-painting methods describe the evolved techniques of fresco and al secco painting. The process, which has been used from Roman times to the present day, is more or less as follows:

• Preparation of the wall. 

The wall, whether of stone, fired brick or mud brick, was roughly rendered with a thick lime plaster containing fillers of fine earth, straw, animal hair or grass. When the layer was dry it was covered with a thin lime plaster (arriciato), which was applied in order make the surfaces as smooth as possible. The wall was then ready for painting.      

• Buon fresco painting. 

On the dry, second coat of plaster the painters first drew a rough sketch of the scene to be painted, usually in red ochre. Next, after estimating how much surface area could be covered in the space of a day's work, they applied a final coat of old, well-matured lime, on which they redrew the lines of the sketch, which had become partly obscured.

Then they mixed the pigments with water or lime water, making a thin wash, which they painted onto the fresh lime surface. If they were unable to finish painting any part of the freshly limed wall, they removed the fresh lime coat, so as not to let it dry in place, and applied a new coat the next day.

• Al secco painting.

The dry, second coat of plaster was given two or three coats of lime, which were allowed to dry thoroughly. When the surface dried it became smoother, because all the pores in the plaster were filled by the lime wash. They then sketched the scene on the smooth white surface and painted it with some sort of brush, using washes of pigments mixed with animal or vegetable glues.

• Preparation of the paints.

The pigments employed in antiquity were inorganic; although it is very probable organic pigments were also used, these would have been destroyed by the passing of time, leaving behind no evidence of their use.

The inorganic pigments are usually oxides, sulphides or other metallic salts. These substances exist in their natural state as minerals. Various other pigments, which do not exist in nature, man succeeded in preparing artificially in his primitive chemical workshops.

Natural or artificial pigments had to be finely ground in a mortar so as to remain in suspension when added to water. They may also have been sieved to remove the coarser particles, or mixed with water in a narrow-bodied jar, in which the large particles would settle out first and the finer ones afterwards.

For buon fresco painting a little powdered pigment was mixed with lime water, but no more paint was mixed at a time than could be used within a few hours, because the standing lime water solution takes on carbon dioxide from the atmosphere and solidifies.

In al secco painting the colour powder is mixed with a water solution of either vegetable glues (various gums, fig milk, etc) or animal glues (egg-white, fish glue, casein, gelatine, etc.). With these materials the painter can prepare a fair amount of paint in advance and keep it for a number of weeks.

THERA WALL-PAINTING

1.  Wall construction and surface preparation.

They used stones and mud for the construction of external house walls, and mud brick, laid upright, or pisé bonded with timber for internal walls. They incorporated grasses or straw into the mud bricks or pisé earth to improve cohesion. This method, one of man's oldest technical achievements, is still in use today. Pliny mentions that men discovered this method of construction by watching birds building their nests with mud and grass.

The stone built walls were roughly plastered with a mixture of mud and straw to make their surface more even. Because of the fineness of the grass and the disintegrated state of the dried mud, most traces have been destroyed and its presence is difficult to detect. Traces are, however, visible in places where a piece of straw had projected from the surface and left its imprint in the plaster (pl. 1). 

Over the mud surface of the external walls a thick coating of lime plaster was smoothed, varying in thickness from 1 to 10 cms. Where the surface of the stone walling was not very uneven this thick plaster is absent, and it is never found on the internal walls.

On top of the thick plaster of the rough stone walls or the mud surface of the more regular stone walls, and over the internal walls, a thinner plaster, made from a more refined lime, was laid. Its thickness varied from 0.5 to 1.5 cms.

Both layers of plaster, in addition to the foreign earthy admixture, sometimes also contained hair and fine vegetable fibres. From a microscopic examination of the traces they have left - uniform tubular holes - we assume they were hairs rather than fibres.

No further layers of plaster, like those seen in Roman or Byzantine wall-paintings, were observed in the Thera wall-paintings.

2. Determination of the paints.

The basic paints employed in the wall-paintings have been known for the last three years from the analyses of Noll et al. We noticed, however, that there were several kinds of red, blue and yellow and we investigated their composition by X-ray diffraction. At first we were unable to interpret the spectra readings, but the explanation of the spectra became immediately clear when we postulated that the ancient artist might have mixed his paints. The conclusion was striking; the ancient artist was able to play around freely with his paints. He knew how to mix and blend them, how to darken and lighten them. We noted the following combinations:

- Yellow and red ochres.

- Red ochre and Egyptian blue.

- Egyptian blue and carbon.

- Egyptian blue and yellow ochre.

Finally, on smooth surfaces we observed varying concentrations of colour (up to twelve times), showing that lime wash was used to thin down the paint to produce different tones.

3. Painting of the surface

The painting is in all cases done on a surface of thin lime plaster. We were puzzled, however about the actual technique used: was the painting on a fresh or a dry surface? After examining the paintings we made the following observations:

1. On large surfaces the edges of the composition or areas of colour were marked out by string impressions (pl. 2). For the marks of the string to have been left one might conclude that the surface was not dry when the string was pressed onto it. Equally, it could not have been fresh, or else the marks would have vanished within a few minutes. Consequently the lime must have been in an intermediate state, at the point when it was starting to dry. And for the marks to have been so well preserved, it is most probable that the string was pressed onto the surface by some sort of trowel or a piece of wood (e.g. the south wall of the Ladies' Room; Marinatos 1972, pl. F and fig. 2). These string marks encourage the assumption that the wall was covered with lime all on the same day. However, given the large dimensions of the room (4.36 X 2.70 m.) it would not have been possible for it to be painted in a single day, and the conclusion is that the surface was not fresh when it was painted, and therefore the technique used must have been al secco.
2. The preliminary sketch of a subject was generally made in paint, but there are many instances where it was incised (e.g. the Shrine in the West Building; Marinatos 1972, pl. J). A macro - and microscopic examination of the incisions showed that a quantity of plaster had been removed, and this could only happen if the incision were on a dry wall surface (pl. 3), because on a fresh, soft surface it would leave a smooth cut. So, since the preliminary sketch was made on a dry surface, the painting must equally have been done on a dry surface.  
3. There exist coloured walls on which the artist has painted his subject (e.g. the widest wall of the Monkey painting; Marinatos 1972, pl. D), and in order to have painted on a coloured wall the surface must have been dry. In this case we assume that the artist gave two or three coats of lime to the surface to make it white, and then painted over the white.
4. That the painting technique was al secco is further supported both by the over-painted parts of many compositions (pl. 4) and by the flaking of the paint, especially the blue. The flaking indicates that the adhesion of the pigment to the wall was destroyed when the bonding medium decayed. We believe that the impossibility of tracking down the organic bonding material is due to its having decayed in the long course of time that has elapsed and not to its never having been present. After all, when we consider that whole timber beams weighing twenty or thirty kilos have vanished, the disappearance of a few milligrams of bonding medium is hardly surprising.

All of the above suggests that the technique of the Thera wall-paintings must have been al secco (tempera or water-based). However, there are certain indications which lead us to think that perhaps they employed a mixed technique:

1. On some large areas of red the paint is firmly bonded to the underlying surface and is abraded but does not flake off.
2. In the excavations fragments of wall-painting have been found stuck face up one on top of another (pl. 5). These pieces are so solidly stuck together that the only explanation is that fresh lime was used to stick them. In such a case, then, the ancient painter covered the area with lime and painted on the new surface. This kind of thing can only occurr with buon fresco painting. In al secco painting one coat of lime would suffice.

Nevertheless, in the Monkey painting (Marinatos 1972, pls. 91 A and B) plaster was used to fill the wall. Perhaps, then, here too we have to do with plaster used for the same purpose, and in fact the technique of Thera wall-painting was not a mixed one, but simply al secco?

----------------------------------------

------------------------------------