THE IN SITU TREATMENT OF THE WALLPAPER IN THE STUDY OF THE LONGFELLOW NATIONAL HISTORIC SITE
Thomas K. McClintock
ABSTRACT—The conservation of a room with the wallpaper painted one color is discussed in the context of a continuing project at a historic house. The treatment problems of such a wallpaper are contrasted with those of patterned wallpapers. The wallpaper was surface cleaned, mended, readhered to the wall at areas of detachment, and losses were filled and inpainted. The advantages and disadvantages of several treatment procedures and materials are discussed.
HENRY WADSWORTH LONGFELLOW (1807–1882) lived in the house on Brattle Street in Cambridge, Massachusetts, from 1837 until the end of his life. Built in 1759, the house remained in the Longfellow family until 1973, when it became a National Historic Site under the direction of the National Park Service.
As part of a continuing project, the house and its contents were undergoing conservation treatments. In the study, which continued to be furnished as in Longfellow's time (Fig. 1) the furniture, curtains, decorative objects and books had already been cared for. It was planned to replace the carpet with a reproduction. After the interior trim was repainted, attention was given to the wallpaper. Since the wallpaper was originally painted a single color (and repainted at a later date), one alternative was simply to remove it and paint the walls to match the original color. Because of the wallpaper's archival value to the room and its good general condition, the decision was made to conserve it in situ. It was treated in the fall of 1979 by the Northeast Document Conservation Center.
Overall view of the study, Longfellow National Historic Site, before treatment. Photograph by Richard Frear, courtesy of the Longfellow National Historic Site, NPS.
The continuing conservation project instituted by the National Park Service has included earlier treatments of other rooms. The printed wallpapers of the parlor, a floral motif circa 1840, and of the dining room, a foliage design circa 1870, were treated by the students of the Winterthur/University of Delaware Program in conservation.1 The front and back hallways, originally papered in 1859 and later painted on numerous occasions with a variety of colors, were painted again in 1950 to match original samples of wallpaper found behind a bookcase. Reproduction paper was prepared for an upstairs bedroom when the original wallpaper was discovered under modern additions. These different approaches to the treatment of wallpaper in different rooms illustrate, even within a single historic site, that a variety of treatments can be developed based on historical information and the determination of condition by the curator and the conservator.
The study of Longfellow House was originally used as a parlor and is on the first floor in the southeast corner of the house off of the entry hall. Three of the walls are covered with a wallpaper painted one shade of brown and the fourth wall is covered with wooden paneling painted white. Along the top of each wall is a wooden cornice and along the base is a wooden chair rail, both painted white. There are two exterior walls with windows and two interior walls with doorways. A total surface area of approximately 240 square feet was covered by wallpaper. The wallpaper is made up of two layers, a thicker wove paper on the outside adhered to a thinner wove paper. This was installed in 1869 over bare plaster and remnants of a block-printed wallpaper, crica 1843, and painted dark brown. At this time strips of bamboo were nailed around the edges to give the room a flavor of the Orient. In 1945 the paper was painted a lighter brown.
Curtains hang from metal bars above the window frames. Against the walls rests furniture; hanging from the walls are several framed items. There is one heating duct on the floor, and ultraviolet light absorbing plexiglas covers all of the windows.
The problems affecting the wallpaper had a number of sources. The underlying plaster was applied in several layers. It was subject to stress as it responded to the seasonal fluctuations of temperature and relative humidity, to the loads and vibrations of human occupancy and to the settling of the structure. Consequently, in many areas the plaster had separated from the wall and slipped away, leaving holes, with fragments lodged under the wallpaper. The most disturbed areas were at the top of each wall where the last layer of plaster abutted the bottom of the cornice and above the windows where the projecting curtain bars were subject to great loads. It was on this insecure base that the layers of wallpaper were laid.
The wallpaper itself was very brittle, with dried mounting adhesive on one side and two layers of paint on the other side, but it had in general accommodated the stressed imposed on it with remarkable flexibility. The shifting plaster caused the paper adhered to it to move as well, forming draws at the intersections of the walls and just below the cornice. There were areas where the plaster layers delaminated, leaving the outer layer of plaster to be supported in place by the wallpaper, and where the plaster cracked and fell away, leaving the paper with no base to which it could be readhered. A lesser problem was the simple loss of adhesion of the wallpaper to sound underlying plaster, detectable by lightly tapping the wallpaper. There were numerous tears and losses in the paper. Most occurred before the second painting in 1945: those areas of exposed plaster were covered with the same surface layer of brown-mauve paint as the wallpaper. There was variation in hue and intensity as some areas had been more protected from soiling and exposure to light. Some areas were missed in painting the second layer and the original darker brown is revealed here. The second layer of paint, an ordinary oil-base interior paint, was thickly applied and brush strokes were apparent. Both layers were well bonded to each other and to the paper support. There was little evidence of cleavage.
The house was heated by a coal furnace until 1974. As a result, wall surfaces were covered with a thick, dark coat of grime.
The objectives of the conservation project were to improve the appearance of the wallpaper and to promote its sound physical condition. It was proposed to clean the paper as possible, to repair areas of damage, to reattach areas of detached wallpaper, and to fill and tone areas of loss. After an examination for condition, at no point was anything but an in situ treatment considered.
Loose surface dirt was wiped off with a soft brush. To remove more stubborn grime, bags of eraser powder were rubbed over the surface, but these were quickly torn by abrasion against the rough surface. Where the surface was determined durable enough, cleaning continued with kneaded erasers. These removed the most grime, produced little streaking if used evenly, and left little residue to clean up. This cleaning was the most important factor in the improved appearance that was a result of the entire treatment.
To reattach the paper to the plaster underneath, a 2:1 mixture of wheat starch paste and methyl cellulose (4% solution of the product available from TALAS) was used, diluted to the consistency of heavy cream. This provided the desired tack, optimum working time, and did not cause stains. At tears or where the paper had separated from secure plaster, the wallpaper could be easily reattached by injecting the adhesive behind it and applying pressure with a wad of cotton for several minutes, which also removed any excess adhesive. At areas of delamination without tears, it was necessary to first make a small hole so that the syringe could be slipped behind the brittle paper without shattering it.
Where there was no firm base to which the wallpaper could be readhered, several approaches were taken. Because the paper was so brittle, it was difficult to pull it away where torn to reach fragmented plaster underneath. Since the paper was of a single color and without a pattern, it was impossible to make inconspicuous incisions following areas of design to lift the paper and repair the plaster. Instead, the plaster was consolidated with a chalk/gelatin (4% solution) putty the consistency of mayonnaise, which was injected with a syringe. The paper was pressed lightly against the putty so that it could conform to the back surface of the wallpaper. This filled the gaps and readhered the loose plaster. After the putty dried, the wallpaper was readhered with paste to this more firm base. Wherever possible, tears were mended by slipping pieces of pasted mulberry paper behind them so that the tear was bridged to provide reinforcement.
Because the plaster underneath could not be made completely smooth behind the wallpaper, particularly just below the cornice where there were large gaps behind the paper, the best solution was to reinforce the wallpaper by inserting paper patches. Where there were large draws, it was necessary to make several compromises. If the draws were torn or were slight enough to be pressed flat, they were treated as above. When the draws were larger and not torn, efforts were made initially to relax the paper by injecting a more dilute adhesive behind it. But upon applying gentle pressure to flatten the draws, the paper would still fracture. The paper was then left untreated with the idea that sound paper, if drawn, was preferable to a surface that was broken up and set down. Fortunately, a large bookcase hides the corner where this is most apparent.
Most of the losses were filled with a chalk/gelatin putty as used in painting conservation. Inserts of okawara paper were used in large areas of loss where the plaster had been painted over but remained rough and unsightly. Otherwise, most losses could be filled evenly with putty and sanded smooth. All of the losses were inpainted with Liquitex acrylics. This medium had enough body so that the surface could be textured to simulate that of the painted wallpaper.
The times spent on the project were: 24 hours for surface cleaning, 43 hours for mending and 26 hours for filling and inpainting, totalling 93 hours.
The conservation techniques described here were adapted for use on the type of wallpaper found in the study of Longfellow House: a heavily soiled, brittle paper without a design motif, painted one color, adhered to a surface that was very uneven in some places. For other types of wallpaper some of these techniques would not necessarily be advisable, such as surface cleaning with erasers, or filling losses with a putty prior to inpainting. Similarly, there are other techniques which were not used on this project which a different type of wallpaper or plaster surface may have permitted or warranted, such as lifting a flexible paper to attend to insecure plaster underneath, or removing the paper from the wall to treat it in a workshop.
I was grateful for the preliminary condition reports prepared by Anne Clapp, for the guidance of Mary Todd Glaser, for the assistance of Irvin Weaver and for the hospitable cooperation of the curator, Kathleen Catalano, and the staff of Longfellow House.
Price, Lois Olcott and Schulte, Elizabeth Kaiser, “Conserving the Paper Stainers' Art: the In Situ Treatment of Two Historic Wallpapers,” Papers Presented at the Art Conservation Training Programs Conference, 1979, Cambridge: Center for Conservation & Technical Studies, Fogg Art Museum, 1979, pp. 5–24.