Detail showing an area with residual starch-based facing adhesive on panel one of the triptych The same area is seen after starch removal. Note that the pattern of cracks seen in the starch has been transmitted to the underlying paint film
The full-scale treatment of the Museum's triptych began with an intensive study of the surface condition aided by the stereomicroscope. The examination, and numerous test cleanings in small areas, revealed that dirt and discolored varnish were conspiring to cast a grayish veil over the painting. The varnish coating, though thin and judiciously applied, was saturating the paint, falsifying and heightening the contrast between some areas by rendering certain colors richer and deeper in value. The image was further obscured by areas where residues of wax-resin and facing materials from the 1959 restoration had not been removed. The adhesive was also beginning to crack the paint film itself. The conservators noted that the residual adhesive was well bonded to the paint, and surmised that as the starch-based adhesive expanded and contracted with normal changes in relative humidity, the paint was tensing and relaxing along with it. This eventually led to cracks, which were observed in local areas of the paint film.
Select for enlarged view of Water Lilies
The conservators undertook cleaning tests on tiny areas and found that removal of the accretions remaining from the facing's adhesive was complicated by the unusual sensitivity of the paint surface to traditional cleaning techniques. Microscopic paint fragments were being picked up even when areas were cleaned with water. Did Monet, like some of his contemporaries, occasionally use a soluble, water-based medium? Or was there some agent on the surface of the painting causing the paint film to break up?
In addition, the examination revealed many problematic areas on all three panels such as inpainting (done by a conservator to replace lost original paint) that had become discolored and paint loss and abrasion, the cause of which was unclear. Identifying the problems was not always easy. Conservator Carol Stringari recalls, "We saw areas of starch that were very similar to areas where Monet utilized underlayers of white, for example, so it was a bit confusing: what areas were starch, what areas reflected his technique? Why did the test cleaning pick up colors in some areas and not in others?"