Bulletin 2 (I:2), December 1963

Home
Français
Introduction
History
Annual Index
Author & Subject
Credits
Contact

From the Laboratory of the National Gallery

by Nathan Stolow,
Chief Conservation and Scientific Research Division

Pages 1 | 2 | 3 | 4 | 5

The Toilet of Venus

By Boucher

In the latter part of 1961 this painting, from the collection of the Nationalmuseum, Stockholm, was circulated in Canada as part of the Héritage de France exhibition. (19) On arrival in Montreal it was discovered that certain areas of paint were in danger of falling away, and as the painting had never been lined since 1746 (its date of execution) there was every reason to suspect that the canvas was in a very delicate state of preservation. It was also probable that it might not stand up well to the hazards of transportation during a Canadian winter. The exhibition organizers, in consultation with the Swedish authorities, decided that the best course of action was to transport it under careful supervision to Ottawa where it could be examined at the National Gallery, with the view to carrying out whatever conservation treatment would be deemed necessary to ensure the safe return of the painting to Stockholm. This, of course, presented a unique opportunity for the laboratory of the National Gallery to carry out documentation, examination, and conservation treatment on a painting of very great merit. During the winter and spring of 1962 the painting was examined in minute detail to ascertain its true condition. The painting was photographed in black and white, in colour, by ultra-violet fluorescence, infra red, and radiography. In addition, a study was made of the palette of Boucher by means of cross-sections of minute fragments of paint removed from along the folded edges of the painting.

The painting had never been lined. The stretcher dimensions of the painting were 38 1/2 by 50 3/4 inches, and it was observed that as much as 3/4 of an inch of the original painting was folded over this stretcher on all sides. At an early stage of the examination it was considered important to reclaim as much as possible of the previously hidden edges of the painting by constructing a larger stretcher, and by modifying the sight dimensions of the frame. The incompleteness of the composition of The Toilet of Venus suggested that the original painting was somewhat larger than that now shown. No documentation has yet been discovered which reveals the original composition. (20) Examination of the folded edges of the painting, and the surface of the painting proper, allowed one to establish areas of overpainting, particularly along the upper edges. This has been detailed in the study of blue sky colour (Figs. 15 and 16). By combining the evidence of radiography (Fig. 13), infra-red (Fig. 14), ultra-violet photography, and microscopic studies, one was able to arrive at a fairly complete assessment of the areas of restoration, repaint, and regions of fragility. A typical structure of Boucher paint from the folded-over edge is shown in Figs. 17 and 18. On the basis of the examinations carried out it was possible to establish a method of treatment: e.g., lining (using the hot table to consolidate fragile paint and canvas, especially loose paint film), removal of the yellowed layer of natural resin varnish with solvents, applying lined and cleaned painting to an enlarged stretcher, inpainting where necessary, and finally, the application of a suitable resin varnish coating. (21, 22) The following are brief notes from the treatment report:

The entire surface of the painting was faced with wet strength tissue paper using dilute parchment size. The tacks were removed, and it was then found that an old repair 1/2" diameter at upper edge 21" from right was firmly glued to the stretcher member. The only manner to free the support from its stretcher frame safely was to cut away the part of the stretcher affected and to chip away the wood from the canvas surface. After filling tack holes with gesso (calcium carbonate and dilute parchment size), as well as other losses, facing paper was applied, to these folded-over edges. The back of the support was vacuumed, the old paper patches removed and all knots: and other imperfections sanded level. Impregnation with wax-resin adhesive, 4 parts beeswax: 1 part damar resin 1 part gum elemi, was carried out on the vacuum hot table. The new unbleached linen lining canvas, warp 32, weft 27 double threads per inch (similar to that of the painting) , was stretched on a temporary frame, moistened with water, allowed to dry, and then restretched. The lining canvas was then impregnated with the wax-resin composition on the vacuum hot table. The lining was carried out on the vacuum hot table, using an underlayer of wide blotting paper to prevent imprinting of the texture of the canvas on the paint surface when vacuum pressure was applied. (23) Time and temperatures were recorded graphically. The duration of the heating process was approximately 40 minutes, with a maximum temperature of 75 degrees centigrade. The entire heating and cooling cycle lasted approximately 70 minutes, and the vacuum pressure measured at the edges was in the vicinity of 2/3 of an atmosphere.

After lining the facing tissue was removed using a damp cloth, and excess wax on the paint surface removed with an emulsion of petroleum hydrocarbons and water. The gesso fills were also levelled at the same time. The cleaning of the old discoloured varnish was carried out with the solvent mixture of equal parts of isobutyl alcohol, diacetone, and shell sol 715. (24) Acetone was used in some areas where the resin appeared to be more stubborn, and in places of overpainting, e.g., upper left sky, right hand and right leg of Venus, left shoulder and head of putti at right. No attempt was made however to remove the overpainted drapery, 4" x 2" at upper edge, 16" from right hand margin, and the later paint, 1 1/2" x 2" at upper edge, 30" from right.

In order to conform with the new dimensions of the painting, brought about by the reclaimed margins, a new stretcher was made of basswood. The new dimensions were 40 1/8" x 52 1/2", compared with 38 1/2" x 50 3/4" for the original one. The lined painting was fixed to the new stretcher with tin-plated tacks.

A coat of damar resin in xylene was applied before inpainting. A grey water colour priming was added to all gesso fills and pigment ground in normal butyl methacrylate and xylene was employed for inpainting. The final coating was carried out with normal butyl methacrylate in xylene sprayed on after two weeks drying time.

Oak strips painted dull black were screwed to all four edges of the painting, and projected 1/8" forward from the picture surface. This provided a protective edge as the picture rested in its frame. The rabbet of the frame was adapted to the new stretcher dimensions by carving out approximately 1/4" after the previously installed 3/4" liner was removed. The picture was secured in the frame by placing a 3/4" raised rim on the frame back to accommodate stretcher thickness (1 1/8") and screwing on a 1/8" thick sheet of tempered masonite as protective back.

Next Page | Other Investigations

1 | 2 | 3 | 4 | 5

Top of this page