Materials,
Technique and Conservation
Shrikant Dabhade
|
||||||||||||||||||||||||||||||||
|
In Western India more than a thousand caves exist of which Buddhist caves at Ajant¡ contain finest wall paintings. These paintings have been exposed to extreme climates in all respects, air, rain and heat, However, they still retain their grace and incomparable technical excellence. Geologically, Ajant¡ area is a part of the Deccan Trap. It is a thick pile of lava flow, interrupted through fissures. Often the upper part of the flow is highly vascular with some large cavities filled with secondary minerals like zeolites and quartz. The caves and the sculptures appear to be executed predominantly in the massive basalt flows. These flows are very hard and resistive to incision. Considering the limited availability of tools in those days, the excavation of tons of rock materials itself must have been a time consuming and tremendous job. This might have been mainly the reason that the work continued for several centuries. The
destructive influence of the tropical climate makes a
painting much less durable, especially on stone. The
conductivity of the base provides a long range in the
temperatures and thus the bond between the plaster and the
stone becomes weaker and weaker in course of the time. The
survival of the paintings against the odd conditions
mainly depended on the severity of the conditions.1
Various techniques of wall paintings In general wall paintings may be classified into four classes, viz, tempera, fresco, encaustic and oil painting. Mixing the colours with a medium, a binding material, which is usually water-soluble like, glue, gum, etc. is usually referred to as tempera painting. Fresco painting is the one in which the colours are laid down without any binding material while the lime plaster is still wet. This is called in Italian terminology as "fresco " (i.e. painting on the fresh). Encaustic painting is done with wax colours which are driven into the plaster by application of heat. Lastly the oil painting technique, which came into prominence in the Renaissance period, has drying as medium. An excellent treatise by Balasaheb Pant Pratinidhi discusses in detail the technique of painting at Ajant¡. He calls these paintings as "secco fresco".2 The technique may be understand by considering four factors, mainly (i) The carrier which supports the ground, (ii) the ground on which the paintings are executed, (iii) the pigments used in elaborating the designs and (iv) the binding medium. The last one is really important in considering the technique of wall painting.
Carrier or support is the actual base on which the paintings are executed. Ajant¡ caves are cut out of hard volcanic trap-rocks, For this reason, the surface of the walls, ceilings and sculptures are full of uneven spherical cavities created due to the evaporation of trapped gasses during the eruption. The walls and ceilings of the caves are the "support", serving as foundation of the paintings to satisfy the requirements of firmness and durability. They are of amygdaloidal trap, which is much weathered. The rough surface provides the tooth for the plaster to adhere firmly to the walls and ceilings. The hard non-porous trap has eliminated all chances of salt efflorescence, and the paintings are free from them. While excavating the caves, the surface of the rocky carrier was further roughed out with chisel creating various types of marks in different degrees to make better bonding and to suit the ground for painting. Three types of rock surfaces have been found at Ajant¡, most uneven, moderately even and even. The most uneven surfaces can be seen on the ceilings and the walls of the cells in the vih¡ras. Moderately uneven surfaces were made on the vertical sidewalls of the vih¡ras of most of the Mah¡y¡na period caves where j¡taka stories are depicted. Unevenness was not maintained in all cases. To bring out sharp details in sculptures and ornamented pillars, high degree of evenness was felt necessary. There is exception though that the surfaces in oldest Caves, IX and X, of the Hinay¡na group were more evenly chiseled.
Nature of layers under the paintings There are several examples at Ajanta where the plaster has come down simply because the surface of the wall was unable to support the excessive weight of the plaster and band of clay. At places, moisture has percolated into the caves from the top softening the earth plaster and thereby damaging the paintings. The plaster had two layers, rough and fine, which are not clearly separated on account of a good binding medium between the two. The principal material used for plastering is sand and lime. The former acts as an inert material. The thickness of the layers vary depending on the surface of the wall. On other sites, it is observed that the finishing of the surface is done efficiently, even polished with the trowel and polishing stone. This is seldom done at Ajanta. The ingredients of the plaster vary in both layers. A survey of the carrier at Ajant¡ reveals that the painters preferred to prepare the first layer with ingredients such as paddy-skins, fibres, and grass seeds. Stone chips were avoided to reduce the weight of the first layer. The quantity of clay was also kept minimum. These ingredients constitute 75% whereas the clay was about 25%. The thickness of the first layer is between 1 to 1.5 inches. The second layer was made of fine clay and sand mixed with fine fibres; at some place cotton and long grass have been found. Sometimes, like in Cave I, paintings were executed on this layer after a fine wash. This layer is on an average 0.25 inch thick. The top layer is made of fine sand and clay. Sculptures and pillars were made to attain high degree of smoothness.
Preparation of wall surface as mentioned in ancient literature The preparation of the wall surface and colours for the painting is given in Vi¿nudharmottara. Part III, ch. 40 verses 1 - 30.3 & 4.
Before starting to paint, vajralepa (hard layer) must be applied, and the instructions for this are given in Abhilasit¡rtha Cint¡ma¸i in verses 141 - 147.
It would also be very useful to take help of the old Buddhist literature. The summary of the Rhys David's translation of the text is as follows:
The ingredients used in the preparation of the plaster for the stone walls, include: (i) brick powder, clay and sand for solidity (ii) paddy skins, fibres, grass seeds, cotton fibres and other organic materials as reinforcing material and consolidation of the layer (iii) clay for plasticity at the time of application (iv) saffron oil for fragrance (v) bees wax oil and mustard seed oil to make it water insoluble and smooth (vi) molasses and liquorices are sticky and crystallize after drying (vii) mudga (Phaselous Munga) after keeping in water for long periods, the protein gets hydrolised which acts as adhesive (nowadays in surface coatings, 72 % of the vegetable adhesives as proteins derived and isolated from soybean, groundnut and cottonseed cake are being used an excellent binders) (viii) grinding with banana paste has similar effect (ix) gelatin obtained from the boiling of buffalo hide, bones or muscles have remarkable property of coagulating, adhesive strength and protecting capacity, and (x) neem extract and extract of wood apple provides excellent protection from insects. The method of preparation of vajralepa gives another clue to the cause of long life of the paintings. The treatment of buffalo's hide, bones or muscles in boiling water, and the buttery substance obtained can be nothing else but gelatin. The
formation of gelatin from the collagen would be as
follows: Even today, gelatin is principally used in industry as binding agent, and number of glues can be made from different sources, which are named after them like, animal glue, bone glue, fish glue and so on. The unique property of gelatin in vajralepa must have played important role in the durability of these remarkable frescoes.
Scientific examination of the paintings It is interesting to understand the technique of painting process at Ajant¡ by the results of scientific investigation done in 1936/37 by Paramsivan 5. A series of experiments was conducted to study the nature of the ground. Micro section was studied after cutting vertically through the different layers composing it. The edges were cleaned from dust and with travelling microscope the section was obtained from the ground which consisted of various layers; painted stucco and rough plaster 9.3-54.3 mm and fine plaster and paint film 0.1mm. The rough plaster was further analysed for the size of the particles. For this the Robinson's method was used. The proportions were as follows. Three sizes were considered < 200 mju, 100-700 mju and > 700 mju. The analysis indicated larger proportion of larger size of particles in the rough plaster. The rough plaster was also analysed for the chemical ingredients namely, silica, iron and alumina were in large proportions than the percentages of combined water and organic matter. This indicates larger proportion of clay used.
The analysis of the binding media is difficult since these are not well-defined substances. The problem of their identification is complicated by three facts: (i) the amount of the medium required to fix up the pigment is very small, (ii) even if the binding medium remains in situ, it changes its chemical character in due course through the chemical action of moisture and atmospheric oxygen, and (iii) the remuneration of the chemists. The tests carried out with the plaster with methylene blue, methyl violet, aid green or iodo-eosine imparted no stain, which indicated that no organic binding medium like drying oil, glue, albumin or casein was added to it. Hence the organic matter reported in the examination in the rough plaster entirely is due to the vegetable fibres and paddy husks which had been specially added to it.
Another aspect of the painting technique is the selection and preparation of the pigments. The Indians have, like other nations recognized three primary colours, red (rakta), yellow (pita) and blue (Kajjal), but added to these are white (sita) and black (¿y¡ma). These five are considered to be pure colours (¿uddha var¸as). The permanency of the pigments depends on their chemical nature. Most of the pigments were inorganic, like iron oxide in various forms, calcium carbonate, various forms of coloured earths, mixtures of oxides and silicates, which remain unaffected. Carbon in the form of lampblack is comparatively permanent. The paintings were not exposed to direct sunlight due to which the lustre remains unaltered. Pita var¸a dh¡tu or yellow ochre is a kind of earth obtained from nearby hills or certain rivers. It is dug out, washed several times with water, the clods broken up and ground nicely in the mortar. By pouring water on the powder and stirring, the coarser grains settle down at the bottom. The water containing the fine sediment is transferred in another vessel where it settles down. This process is repeated several times till all the fine powder available is recovered from water. The supernatant liquid is poured off and the yellow mud is applied to a new earthenware vessel so that it absorbs the moisture. The yellow cake so formed is broken in small bits and preserved for use. The yellows are of four kinds, namely, ¿var¸a that is the colour of gold; kapis¡, which resembles the rajanisira; pita, which is the colour of haritala (chrome yellow) harita, which is that of the cat's eye. Rakta dh¡tu (gairika) or Indian red is also an earthy substance found on hills, and the treatment is similar to that for pita dh¡tu. Red lead (¿ind£ra) is soaked for half a day; orpiment (manassila) which is arsenic sulphide must be powdered dry and soaked for five days and then ground with water on the sixth. The four variants in the red colour are: aru¸a, which an be recognized as that of blood of the hare, rakta, which resembles the red of shoe flower; son¡, that of the kimsuka flower, which itself is the colour of the beak of the parrot, p¢tala which is the colour of lac dye (lakÀarasa). The blues are four in number: n¢l¡, which is the colour of the clouds; ¿y¡ma, which is of the crow; karala which resembles the colour of the neck of peacock; k¤À¸a, which resembles the jet-black wings of the beetles. Gold colour: pure gold is beaten into very thin leaves which are cut to small bits, put into mortar, a small quantity of fine sand is added and both are ground together till gold is reduced to very fine dust. The paste then is levigated and several washings done to remove the sand particles. Black was prepared in the following manner. Good oil is put in an earthen pot with a clear wick, another earthen vessel cleaned inside with cow dung powder and put over the flame of the wick; the carbon black is collected and washed, then dried; later it is finely ground on a metal mixed with water and wood apple juice. Here there is doubt about the metal used (manganese?). White colour is prepared from the slaked lime, conch, mother of pearl, shells or chalk / white earth. This fine powder is put in mortar of wood or granite that is not liable to discolour. This is mixed and pounded well into a paste, treating the same with milk of tender coconut. The paste is dissolved in warm water, filtered, and the solution allowed to precipitate and settle. The mixing of pigments to form other colors is called upavar¸a. White and black in equal quantities give colour of the elephant (gaja-var¸a); red and yellow in equal parts give the colour of fruits of vakula tree; orange colour, if twice the quantity of red is mixed with one of yellow it yields colour of deep orange (ati-pita). A mixture of equal quantities of haritala (chrome yellow) and ¿y¡ma var¸a (blue) make a fine green resembling the green of parrot. Thus by mixing larger number of pigments, much larger variety of mixed colours (sank¢r¸a-var¸a) can be obtained. In all the cases, the pigment before application is mixed with gum of the neem tree, or gums of the wood-apple tree. Glue is prepared from the boiling of the buffalo fresh hide, and mixed till the paste assumes consistency of butter, which then is rolled into balls and dried. Whenever required, this is dispersed in hot water ready to use.
The various texts describe various methods for application of colours. Various instruments like crayons, brushes have been recommended. The crayon, are vartik¡, kitta-varti or kitta-lekhan¢. The lampblack was ground with little boiled rice and rolled in the shape (varti) of middle finger (karnik¡), which is called vartika or crayon. The brushes or lekhan¢ required for painting were classified as fine, medium and large sizes (sukÀma, madhyam and sth£la). The handles of the brushes were made six angulas and the hair of the brushes six yavas in length. For sth£la lekhan¢s, the hair in the ears of the calves was used, for madhyam¡; the hairs of the belly of goats and, for fine ones the hairs of the tail of squirrel. A metallic nail was driven in one end of the handle; the hairs were secured around the metallic nail either with lac or with fine thread. In the texts, it is recommended that the artist must provide himself with brushes of three different sizes made with three different hairs. The sweeping (lepana) is done with coarse ones.
Analysis of pigments at Ajant¡ The following pigments were identified from the samples of paint collected from Ajant¡ caves.6. It was not possible to collect all the pigments from one and the same cave lest there should be damage to the paintings by collecting large amounts of sample.
Condition
of the murals and conservation The internal causes of deterioration of the murals are the development of cracks, decay of binding materials, chemical changes in the pigments, fading of colours, friability of the plaster, movement and bulging of plaster, flaking of pigments and priming. The last two being caused by the violent changes in humidity and temperature. It
is very interesting to note that these caves
remained unattended by the experts for almost
one century, who earlier had primitive methods of
cleaning. Initially, Sir J. Marshall was kind enough to
consult the three most notable experts of
"restori" in Italy. It is reported that Cecconi
favoured very dilute caustic soda and few drops of
hydrochloric acid.7
They fixed the paint by solution of shellac
dissolved in alcohol, dammar and turpentine, This
varnish turned yellow after decades. Between the years
1920 and 1922 the experts Cecconi and his assistant Count
Orsini used the following methods for fixing the patches
of the paintings: (a) injection of casein-lime, where
the gap was narrow, (b) filling plaster of Paris or
lime, fine pazzolana, where the cavity was large, (c)
fixing nails in dangerous parts for supporting the coarse
plaster, (d) in some places strengthening the rough
plaster by sticking strips of linen on the surface with
hot gelatin and removal of linen by means of hot water
after the parts have been secured to the wall, and (e)
liberal use of unbleached shellac dissolved in alcohol for
general preservation. During the later years the same technique
was adopted by the
Department of Archaeology of Hyderabad. Archeological
Survey of India later took up the conservation work from
1953 onwards.
Treatment of wall painting surfaces during the last five decades10 The nature of the treatment must be different from place to place since the technique of painting underwent changes though various generations of the artists working at the site. It is the experience, which guides the conservationists working at the caves. Due to the limited number of experts available the work cannot be taken up on war footing. It is also important to note that complete chemical analysis of the paintings is not going to solve the problem completely. The important thing is to preserve the existing ones and stop further disintegration. The conservation of Ajant¡ paintings is a continuous process. The treatment also varies from place to place depending on the nature of the damage. Since the paintings have not been done at one time, the composition and problems are varied. During the five decades scientists have arrived at some methods. These methods deal with various aspects of defects like those relating to the structure, optical, relating to the appearance, and lastly relating to decay, decomposition, and disfiguring due to environmental and biological factors.
The first problem in conservation is the consolidation of plaster. The painted stuccos are often loosely holding on to the ceiling in such a manner that a mere touch was enough to detach them. It is considered necessary, therefore, that these loose stuccoes are consolidated and strengthened even prior to taking up the problems of general cleaning and preservation. The edges are then filleted with plaster of Paris suitably coloured to a neutral tint by addition of yellow, red ochres and lamp black. Thus the painted surfaces are saved before being taken up for cleaning, consolidation and preservation under a comprehensive scheme. The dust, dirt, cobwebs, insects, nests, insect cores and insect waxes are removed by gentle brushing and through the use of organic solvents. Such excretions as soot, lichen and moss are cleaned with rectified spirit containing few drops of ammonia. The cleaning in most cases is carried by aqueous solution of ammonia with surfactant like teepol. After cleaning topcoat of polyvinyl acetate of suitable concentration is given for protection. For removing the hardened varnish and other accretions, organic solvents with varying solvent power such as acetone, butyl acetate, butyl lactate, triethanolamine, dimethylformamide, morpholine, cyclohexamine are used in different proportions. Care has to be taken since some of these have disruptive effect on the varnish. Incrustations, which do not respond to the action of the solvents and their mixtures, are treated with a solution of formic acid in alcohol. Bulging on the northeast wall of the Buddha Chamber in Caves 6 and 16 was fixed back by using Japanese tissue paper and cellulosic sheet for wetting to soften. PVA solution in Toluene or Fevicol mixed with plaster of Paris was then injected into the cavities. Bulging surfaces were carefully pressed till the adhesion was complete. Colour matching was done wherever necessary. Removal of copper nails applied half century back in the painting panels of Caves 9 and 10 on the east wall of the Cave 17 were removed with great skill. The areas around the nails were wetted before pulling them out. The gaps caused were filled with a mixture of fevicol and plaster of Paris.
The problem of "chalkiness" or "bloom" sometimes appears on the treated surfaces. Scientists have found reagents for satisfactory removal of dark shellac varnish that obscures the optical values of the paintings. Silicone resin is extremely effective but its subsequent elimination if required was not an easy matter. Bleached shellac was ruled out, on account of tendency to turn yellow, brittle and develop cracks with age. Methyl methacrylate resin is recommended. Persistent research has been carried out by department for softening brittle pigment layers it was found that they could be softened with the help of emulsifying agents and detergents. For removal of chalkiness and remains of shellac varnish organic solvents such as cellusolve, acetone, Butyl acetate, Di-methyl formamide, cyclohexylamine, morpholine etc were used individually and in suitable mixtures. More resistant incrustations, which did not respond to action of these solvents, were treated with a 5% solution of formic acid in alcohol. This led to softening of the upper layers of accretions. Formic acid facilitated the removal of underlying accretions with organic solvents. A coat of 1% polyvinyl acetate on the cleaned paintings was found to impart full saturation to the colours. . The whitish accretion occurring in the resin layer was observed to have contained inorganic constituents like phosphates, oxalates etc. It is likely that these constituents are derived from excreta of bats. White chalkiness was treated with solvent mixture containing Formic acid with good results.
In a study of the biological deterioration of the murals of Ajant¡ and Ellor¡,11, nearly 25 fungal species belonging to different genera was isolated. Shrivastava and Mathur studied the degradation carried by the silver fish.12. Ajant¡ paintings were found heavily infested with silver fish. The silver fish specimens were collected and reared in the laboratory. Other problem besides silver fish is beetles. Lot of forestation is taking back on the hills of Ajant¡, which results in more air borne insects, larva and microorganisms causing small holes especially on the painted surface. The types and the population of the insects change with the nature of plantation done. Biocides were tested on vitro and in situ on replicas against isolated fungi. The efficacy of biocides likes O-phenyl phenol, p-chloro cresol, ceto pyredinium chloride, phenyl mercuric acetate, sodium pentachlorophenate and preventol R-90 have been tested against different fungal forms isolated and identified on Ajant¡ paintings. Different concentrations of biocides were inoculated with the test fungus in a suitable media and their efficiency was evaluated by recording colony diameter at regular intervals. It was found that phenyl mercuric acetate was very effective than other biocides and also prevented growth of fungus but required in higher concentration of reagents. The efficacy of these biocides and their effects on the materials of wall paintings were also studied on test samples of wall paintings prepared in the lab. Effect of temperature, pH, and pigment on the growth of microbes was also observed after a series of experiments. Extracts of custard apple seed, Pongamia (karanja) and neem seeds have been tested to control carpet beetle, silver fish on wooden and paper material. Extracts of deoiled neem kernal and karanja seeds were tested and found to have been effective against insects like silver fish, furniture and carpet beetles. Four biocides viz. Zinc DDC 5%; Menthol 1,2 and 5%; have been tried and tested on replicas of Ajant¡ wall paintings. Various methods have been used for the treatment of insects. It has been observed that the bare plaster on the walls and ceiling inside the cells are the main breeding ground for the insects that were found to migrate to the painted surface. In one instance the spraying of fungicide solution, like pyrethrum mixed with alcohol and kerosene oil was carried out for the extermination of the insects from the cells devoid of paintings. In another case, the application of 0.5% Sodium pentachlorophenate solution for studying the effect of further burrowing of holes by insects has also been tried. A mixture of pyrethrum and phosphorus formulations has also been found to be useful. Once fumigation with CARBOXID gas was carried. In the experimentation17.6 Kg of Carboxide (Ethylene Oxide and Carbon di oxide in 1:9 proportion) was released into the cave. The rate of gas was kept 1.35 Kg/100 cu ft. The cave was sealed for 30 hours. Insect activity was checked with pyrethrum (0.25%), which was sprayed as insect repellent.
Present condition of the paintings Originally, almost all the caves seem to have been painted. Even the sculptures were decorated with the colours. The total area of the paintings at Ajant¡ is approximately 3500 square meters which is very large and requires more human resources to maintain13. On account of fluctuations in the temperature and humidity there is a sudden drop in the velocity of wind and it causes deposition of dust and dirt on the paintings. Besides, the visitors who come in large numbers also bring a lot of dirt. The future of Ajant¡ paintings and their conservation needs a serious thought. The caves are now structurally maintained in very sound condition. To a great extent they are free from water percolation and seepage. The study of various environmental factors of the caves like the periods of growth of insects, Carbon Dioxide and dust content of the air of the caves, coming of large number of visitors, lighting conditions of the caves, variations in the temperature and relative humidity of the atmosphere with the changing of seasons etc, need special attention. Some of these problems have already been tackled by the authorities.
|
