It was the winter of 2007, and within the ROM’s dark, silent storage vaults 97 paintings awaited permanent exhibition in the airy new Sigmund Samuel Gallery of Canada. My task, as the ROM’s new paintings conservator, was to create a plan that would ready this daunting number of works for display.
Art conservators are entrusted with the task of preserving art and artifacts. We do this by analyzing and stabilizing the physical components of a work. Most of us specialize in a particular medium: the ROM has conservators of textiles, paper, metals, stone/glass, decorative arts, and ethnographic objects, as well as paintings. In the past when restoring was the main job of the profession, we were called restorers, but nowadays we not only restore, but also research, analyze, document, and preventatively conserve, preserve, and reconstruct objects of art. The bedrock of the profession is to seek the original intent of an artifact or piece of art—the vision its creator intended the public to see without the sometimes distorting effects of time and old restorations—and to preserve its original integrity with minimal intervention so that it will last into the future.
When we assess a painting for treatment, the methods we use are not unlike a crime scene investigator’s. We make site visits, usually to outline travel requirements to ensure a painting’s stability during transit or to assess the risk of pest or mould infestation. And at the lab, we conduct routine tests and examinations to try to discern exactly how and where past damage has occurred. One of these tests involves micro-sampling: we take minute paint samples for analysis. Another looks at solvent sensitivity: we use tiny amounts of solvents to determine the solubility of the paint and surface coating layers. Tests such as these can provide a better understanding of the material composition of the painting’s image layers—and they also aid us in developing a treatment proposal. (And sometimes they reveal fakes, but that is for another story.) The most revealing examination methods involve observing paintings under various types of light—raking light, infrared light, and ultraviolet light—each of which provides different clues to what may lie beneath a painting’s surface.
When it came to our Canadiana project, some of the paintings were newly acquired, while others were icons of the collection, such as The Death of General Wolfe by Benjamin West and studio. They came in all shapes and sizes, and in varying states of repair. To handle the scope of the project, the ROM hired several full-time and two part-time painting conservators. We found many of the 97 paintings to be in stable condition, but a few, about 15 percent, had tumultuous histories—they had a storied provenance, had already endured heavy restorations, or required a new method of display.
Our tasks on three of these paintings exemplify the detective work involved, the cunning and skill that conserving art can sometimes require. These paintings and the stories and images showing how they were conserved are featured in the exhibition Returned to Former Splendour, now on display in the Wilson Canadian Heritage Temporary Gallery within the Sigmund Samuel Gallery of Canada.
View of the City of Toronto
By Edward Taylor Dartnell / c. 1850 / 105.7 x 186 cm
View the painting before conservation
The ROM team began work on the oil-on-canvas painting View of the City of Toronto by scanning it with UV light. Similar to “black light,” UV is good at revealing surface anomalies not visible to the naked eye. One problem commonly revealed with UV light is overpaint—paint added at a date later than that of the original work that is not on top of a repair, but hides the original painting. In this case, the UV scan revealed discoloured overpaint extending along the entire left edge of the painting. This thick discoloured coating of oil paint was hiding serious damage. We found a huge multi-directional tear in the upper right measuring some 32 x 32 cm (more than a square foot). Flaking paint indicated previous water damage, and telltale signs—transparency of paint layers, peaks of canvas threads showing through, and a scoured appearance under the microscope—indicated past overcleaning. Overcleaning, which occurs when paint strata are eroded by too much mechanical force (friction) or chemical means, puts the paint layers in a delicate condition, an oversensitivity similar to allergies in humans.
Because our treatment had to be extra sensitive, we used the least aggressive solvents that would gently clean and remove the overpaint.
Fortunately, under UV light, the conservator was able to determine with pinpoint accuracy which brushstrokes were painted by the original artist and which contained newer paint, applied to conceal the significant damage. Under microscopic magnification, with great care the conservator applied solvent, and used mechanical methods, such as a scalpel and dental pick, on the individual brushstrokes to remove the overpaint.
The large tear needed to be repaired from both the front and the back of the painting, but a fabric lining restricted access to the back of the canvas. To see the tear from both sides, ROM conservators had to do a “lining reversal,” applying heat to soften the adhesive and remove part of the lining fabric. To protect the “healthy” part of the painting from undue stress, only part of the lining was lifted. Fill material—in this case gesso, glue, and other additives—had been spackled on to cover the tear’s seam. Once the old fill was removed, each thread was painstakingly re-woven and fused together to create a seamless mend. With a method called “thread-by-thread” tear mending, the conservator uses a microneedle, glue paste, and a micro tacking iron to fuse each thread back together. As the work progresses—it can take several hours to mend one inch of tear—individual strands are held in place with acupuncture needles. After new filling was added to hide the mend, the conservator inpainted, discretely adding new watercolour paint, which is easily removable in future if necessary, to cover the fill material and any areas of paint loss. The job was finished off with a coat of synthetic varnish, which doesn’t discolour as older versions did. What had been a poorly repaired area now looked uniform and intact.
By Frank Pebbles 1896, 1897 / 266 x 177 cm
View the painting before conservation
This portrait of a notable Native Canadian physician and businessman had been stored flat and unsupported with no existing frame. It had not been viewed upright in more than 30 years. A painting, especially of this scale, needs to be under constant tension in order for the layers of paint and preparation to remain adhered to the canvas. Without tension, the canvas could warp and the paint layers could separate.
An ongoing dialogue with the curator responsible for the painting is pivotal to the conservation process of almost any work of art. Curators and art historians provide insight into the history of the piece—its context, its meaning, and where it fits in the genre and the artist’s oeuvre. Conservators provide the material history—the nuts and bolts of what comprises the physical painting.
For this artwork, the conservator determined that the painting was oil on canvas and had likely been revarnished at least once in the past hundred years. Most interestingly, the conservator unearthed a “double signature and date” at the bottom of the painting. It appears that the artist painted the canvas once in 1896 and repainted over it in 1897. Conservators can’t necessarily explain why an artist would do this, although from artists’ journals and other documents we know that sometimes the reason is that the artist, the sitter, or the patron, was unhappy with the results and wanted a different, more flattering, or more accurate look. But our diagnostic work can reveal that it occurred, providing an objective framework on which the curator or historian can build or validate theories about the painting.
Another puzzle was that Dr. Oronhyatekha had a crudely cut perimeter and no stretcher or frame. The curator knew that the painting had been cut out of its original frame with a sharp implement before its arrival at the ROM. With this knowledge, the conservator was better able to determine the original dimensions of the canvas, which provided the information needed to construct a new custom-made stretcher.
Before the painting could be placed over a stretcher, a new synthetic canvas backing needed to be applied, enlarging the canvas’s size enough so that no painted portions would be in danger of being folded over the stretcher’s sides. Called “lining,” the process is essentially the reverse of our work on View of the City of Toronto, for which we removed the lining. Adding a lining involves brushing or rolling on a heat-responsive liquid or film adhesive to the back of the original canvas and then laying the painting over a new larger piece of fabric; using heat and constant suction with a vacuum pump, the two parts are welded firmly together. Because of the painting’s large size, the lining had to be done on special equipment at the Art Gallery of Ontario. Their Maxwell combination table, sometimes called a hot table, is larger than the one in the ROM’s lab and has a built-in evacuator to create suction. This equipment works like a giant electric griddle pan to heat the painting. The new backing it attaches to the artwork provides added support to the original canvas and paint layers.
Constructing a new frame for this large painting was a monumental task. An elegant presentation was desired, so the conservator and curator consulted to select a simple frame profile with a toned gold-leaf finish. The frame’s pieces were custom-milled from basswood and then water-gilded using 13-karat white gold. Despite its luminosity, gold leaf is very thin: you’d need 26 stacked sheets of it to equal the thickness of an onion skin.
George Townshend, 4th Viscount and 1st Marquess of Townshend
Attributed to Gilbert Stuart / c. 1786 / 265 x 173.5 cm
View the painting before conservation
The third major project, also an oil on canvas, began as a straightforward surface cleansing of dust and grime. One of the most simple and effective solvents conservators use is saliva, gently rolled onto the surface of the painting with a cotton swab, followed by a fresh swab of distilled water. An ideal cleansing agent, saliva contains a great deal of water, weak surfactants (soaps), and a variety of ionic materials, the chief of which is sodium.
As the treatment progressed, we decided to remove the disfiguring varnish that was dulling richly coloured areas of the painting. For both aesthetic and protective reasons, oil paintings often receive a final coat of varnish once the artist has finished. Varnish provides all colours with an optimal level of uniform saturation while acting as a film that protects the painted surface from the harmful effects of dirt, light, humidity, and minor physical abrasion.
Applied by spraying or brushing, varnish forms a thin, transparent layer on the surface of a painting. Historical varnishes were made with tree resin dissolved in a solvent, and have the unfortunate tendency to discolour and degrade over time—necessitating their eventual removal, usually with solvents, such as ethanol. Since the 1950s dozens of surface coatings manufactured in various industries, the vast majority of them synthetic, have been investigated for use as picture varnishes. In this case we used Paraloid B-72.to replace the older coat of varnish.
This painting contains a compelling example of the phenomenon known as “pentimenti,” ghosted images that can be seen beneath the painting’s top layer. Different from what is seen above, these images signal that the artist reworked certain of the painting’s elements. The original position of Lord Townshend’s feet, for instance, is visibly different from their final positioning. The word pentimento derives from the Italian pentirsi meaning to repent. Three different conditions can lead to pentimenti: the upper layers of paint become more transparent over time; cracks develop in the uppermost layers to reveal different-coloured paint underneath; or different-textured paint below reveals brush marks going in a different direction than the ones on the top layer.
Historically, restorers often covered up pentimenti with excessive coats of paint, usually to hide awkward areas of the painting, or (unfortunately) to exhibit their own creative flair. Today, pentimenti are considered an integral part of the image and history of the painting. Despite the fact that they can detract from the painting if they are very noticeable, conservators are no longer likely to cover them up with overpaint. In this painting, we left the distracting ghosted images as we found them.
The Samuel Gallery now displays these 3 paintings, alongside the other 94 from the storage vaults. While all 97 treatments are complete for now, the conservation process is an ongoing one. Some objects will always need treatment, but the vast majority are under preventive conservation: we aim to maintain proper environmental standards and display requirements.
In essence, a work of art or an artifact is a communication from the past. Viewers are a brief audience to that communication. It is the conservator’s job to be mindful of the painting’s past and to preserve it for the future.