By Ph.D. Biljana Topalova-Casadiego, Painting Conservator

Origin and technique

Munch chose a rough, jute canvas with a definite texture as the base for this painting (fig 2). It has been primed with unevenly applied chalk, and because of this, the rough slub of the fabric is often clearly visible (fig 3 and fig 4). This layer of chalk covers the surface that is painted upon, but not the areas that are stretched over the frame. This indicates that Munch himself could have been responsible for preparing and stretching the canvas. The priming has partly absorbed the binding medium used in the layers of paint, and this contributes to the dry, matte, surface appearance. At the same time, the intensity of the original colour is maintained.ⁱ

The painting has been executed with quick, spontaneous brushstrokes. The various parts of the painting seem to have been worked on in quick succession and there are few overlapping areas of colour. The contours, or sketched outlines, also appear to have been painted directly with the brush. The colours harmonise with the effect created by the texture of the canvas and the priming – in many of the lighter areas, Munch has allowed the priming to show through the layers of paint, or barely covers it with a thin layer of colour.

Nevertheless, the thickness of the painted layers varies. The violet and pale green areas, as well as the brown in the trunk of the tree, are thin and almost watercolour-like. The man’s clothes and hair, as well as the crown of the tree, consist of a fairly thin, but more opaque layer of colour, with darker contours, accentuated by some of the brushstrokes (fig 5). The faces and the woman’s arm are painted with mixture of colours that contains white. Thicker layers of paint are to be found almost exclusively on the woman’s dress (fig 6 and fig 7), where the visible imprints left by a colour-loaded brush, are to be found. Coloured splashes and traces of paint runs can be seen in the areas around the underside of the woman’s dress. This may be due to nonchalance, or demonstrate a desire to allow chance to play a part in the painting process.

The colours are mainly gathered together in fine, curving areas that are clearly delineated in relationship to one another – this leads to associations with work in stained glass. The harmonies are discreet, dominated by an elegant blue-violet tone.

Viewed through a microscope, the samples of paint tell us much about Munch’s use of colour and the ways in which he mixed itⁱⁱ. He used a limited palette, and concentrated mainly on relatively clear, bright pigments. A paint sample taken from the violet background in the upper, left corner shows various layers that contain organic reddish paint, and ultramarine, chalk and lead white paint (fig 8). Layers of cinnabar have probably been used for the blood-red plant to the left of the painting. For the green leaves that have been painted on top of this, chrome oxide was probably used (fig 9). Chrome oxide, with small grains of ultramarine and yellow are also to be found in the green coloured background (fig 10). A microscopic sample from the woman’s hair indicates the use of cadmium yellow and this layer of colour also contains lead white and zinc white and traces of small, black grains. In order to attain lighter nuances in the hair, the artist has applied a thin layer of lead white with occasional grains of black and blue pigment on top of the first layer of yellow paint (fig 11). One eye-catching detail is the use of “gold” in some brushstrokes. This, together with several black lines, defines the structure of the hair (fig 12 & fig 13). The layer of “gold” consists of brass – small flakes of copper (Cu) with approximately 10 – 15 percent zinc (Zn) (fig 14). Similar use of gold is known only in few other Munch paintings, but it tallies with a tendency towards the decorative and the exclusive in art being produced at that time.

Condition

During the first years after Munch’s death, the painting was stored at Ekely. In the 1950’s it was lined on a supporting canvas and remounted on new stretchers. From 1975, dates the forst condition report of the artwork - scrapes in the paint, watermarks and stains were registered. During the 1990’s, the painting was mounted behind colourless, laminated and UV-absorbing glass with anti-reflex coating, as well as a new frame.

In May 2002 the painting was examined in connection with loan to an exhibition in Vienna. The reverse side of the canvas was dry cleaned with Lascaux Wallmaster (100% Latex) and vacuum cleaned. Insecure layers of paint were secured and fibre samples were taken from the canvas at the lower right hand corner of the painting in order to ascertain the presence of lignin. The sample was boiled in distilled water before the test solution was applied. The sample gave a positive result, showing the presence of lignin (fig 15).

On large areas of the surface, flaking is apparent – in particular the side areas and the area in front of the figure of the woman (fig 17). Because of this, the colour of the canvas is visible. This colour has also changed in relationship to the original condition and the areas closest to it (fig 16). When the painting was remounted and lined to the new canvas, some of the thickly applied paint on the surface was pressed flat. In the areas where the coloured surface’s canvas structure is retained, shiny areas are visible as a result. Under the microscope, loose fragments of paint are visible and in areas where more thickly applied layers of paint occur – for example, the woman’s dress – there are places where there is insufficient contact between the coloured layers. Under ultraviolet light, areas of retouching of former flaking are also visible. This has not been registered in previous reports, and many of these areas have changed colour at a later date. (fig 17, fig 18). In addition, new, un-retouched flaking is visible and there is a moderate layer of dirt and dust on the surface of the painting.

Treatment

In order to make an informed decision about the best method of conservation for the recently registered flaking, a series of tests were carried out. These showed that a 2% solution of sturgeon glue applied to the surface with a fine brush was a successful method of re-attaching the paint particles to the canvas. After this excess glue from the surface was removed with de-ionised water and the treated areas could be dried under gentle pressure. Old, retouched areas were left as they are. The stretchers were improved and a backing board was mounted. The outer no original frame of the painting was also found to be too tight, so a new frame was made.

During the conservation process it became clear that the structure of the paint is brittle and even thought the surface has been made more stable as a result of the conservation work carried out, it will be necessary to examine the flaking areas at regular intervals.

Photographs:
Svein Andersen and Biljana Topolova-Casadiego, The Munch Museum

Footnotes:
i V. Jirat-Wasiutynski, H. Travers Newton: Technique and meaning in the paintings of Paul Gauguin, Cambridge University Press 2000, p. 54
ii Laboratory analysis nr. M0042 by RNDr Janka Hradilova (Academy of Art, AVU, Prague) and Dr. David Hradil (Academy of Science, CAV, Prague).