Author: Danielle Creech

Theresa Smith of Weissman Preservation Center, Harvard University presented her treatment of the heavily edited typescript manuscript of Marguerite Yourcenar’s L’Oeuvre Au Noir. The manuscript, begun in 1956, had been bound into 2 volumes after its completion in 1968. The author’s editing techniques presented unique challenges as she made changes to her manuscript by taping new strips of paper over old sentences. Nor did she stop at just one layer. Throughout the course of conserving the manuscript, Theresa found pages with no less than 17 layers of changes, all stuck together with pressure sensitive tape. To make matters even more complicated, Yourcenar would often edit over the typewritten emendations with handwritten notes, making the tape on which these notes were written a part of the manuscript. Long term stability concerns aside, the tape with these notes could not be removed from the manuscript.

When all was calculated and done, the manuscript was estimated to have 6,725 inches of pressure sensitive tape (most of it failing) and at least eight different paper stocks. Tape on the coated paper stocks remained strong and in place, and in the past, researchers with more curiosity than sense had pried up and damaged the strips in an attempt to see underneath them.  On the other hand, the tape adhesives had mostly failed on the uncoated papers, leaving the strips of edited text floating free.  The manuscript was in an extremely compromised condition and  access not been granted to researchers for years. Obviously something needed to be done, but the project was a complicated one that spanned two conservators.  Theresa, who collaborated closely with the collection’s curator when making treatment decisions, worked through the manuscript one page at a time to carefully reconstruct the complex and layered structure of the emendations.

Briefly, here are the primary points of the treatment as completed by Theresa.

• Emendations that obscured text were removed mechanically and hinged in at the spine.   Theresa found wheat starch paste to be an effective adhesive when applied quickly and firmly, even on the erasable bond papers.
• Loose emendations were hinged into place.  If these emendations were on tape carriers, the carriers themselves were hinged into place.
• Staining was not treated, as it often helped reconstruct placement of the loose strips, and could be of use to future researchers.
• Tape that still held strongly to the page was not removed even though there was concern that the adhesive of this tape might creep out in future years, causing more problems.  Time constraints meant that the tape would stay put, at least for now.
• Adhesive residue left behind by any necessary tape removal (only on obscured text) was mechanically removed and coated with cellulose powder to reduce tackiness.
• Handling notes were included in the volume, in the enclosure, and in the card file for library staff.

Theresa’s presentation was yet another reminder that there is no “one size fits all” solution in the conservation profession; flexibility and a good sense of humor are key!

Paula Ricciardi, John K. Delaney, Lisha D. Glinsman, Mathieu Thoury, and Michelle Facini of the National Gallery of Art used several analytical methods to study the media used in the Praying Prophet, a manuscript cutting illuminated by Lorenzo Monaco c. 1410/1413.

Up front, let me say that analytical techniques such as image spectroscopy, XRF and FORS analysis are way outside my balliwack, so please feel free to comment on my post to correct any missing or mis-stated information. My feelings will certainly not be hurt!

So, the crew at NGA used a Si-CCD camera with color-corrected lenses and their specialized set up allowed them to reduce the light levels needed to capture their imagery, thus making the analysis safer and easing their conservator’s mind. They captured several images in the 400-950nm (visible through near-infrared range). They compiled the images into an image cube for analysis, calibrating it to black and white standards that allowed them to gather both reflectance and luminescence information. This allowed them to differentiate between a wider range of visually similar pigments.

Image spectroscopy alone cannot identify a pigment, however, so they also did XRF (which identifies elements such as copper, arsenic and lead in paints) and FORS analysis (which identifies different light absorption bands). One particular example she gave of the the need for both XRF and FORS analysis was the presence of copper in a green area of the miniature. Since copper was a common material used to make green pigment, the XRF identification of copper in that area was not surprising – but FORS analysis showed absorption bands that indicated the presence of azurite. So contrary to expectation, the green was not copper based, but azurite + a yellow pigment!

Even more intriguingly, when analyzing the vermilion robes of the prophet, they found absorption bands that did not match any of the pigments in their database. Curious, they collaborated with a conservator to mock up samples of vermilion pigment in a variety of different binders (egg glair, gum arabic, egg yolk). They found that the mystery absorption bands matched egg yolk. (This is exciting, as they are the first to map any sort of binder using FORS analysis).

Egg yolk was a slightly surprising find, as egg yolk is not well known to be a binder used in illuminated manuscripts. The team was able to find a passage from an early 15th century instruction manual on illuminating manuscripts that indicated egg yolk should only be used as a binder only when painting bodies (and specifically not in writing or when painting flowers). But the team knew that Lorenzo Monaco was also a panel painter (a media better known for egg tempura), so they wanted to dig deeper. Was he using egg yolk because of his panel painting experience? Was he the only one to use egg yolk in illuminated manuscripts? Their analysis found egg yolk was consistently used in the robes of figures throughout the miniature and in other illuminations by Monaco. Most of these robes were vermilion, but even the green robes showed the egg yolk absorption bands. Areas outside figures (such as decorated initials, foliage, etc) did not show evidence of egg yolk, although the robes of figures hidden away in decorative foliage did – suggesting that the difference in use was not just a matter of master painter vs. workshop painter.

When the team turned their attentions to other illuminated manuscripts of the era, they were unable to find egg yolk present, even when the artists’ responsible were also known panel painters. Obviously, more research needs to be done to further explore this area, but one of the most telling points of the talk (for me, anyway) came during the question/answer section. One conservator noted that while the talk did not make her rethink her treatment decisions based on concerns about the items themselves, it did make her think about how her consolidant decisions might muddy the waters for future investigators. This is a very good point, and one I will certainly be mulling over as I consider my own treatment decisions. Although, to be honest, the thought of inspiring some as-of-yet unborn scientist to curse my name is just a bit tempting.

Tammy Jordan of Etherington Conservation Services – East presented on her treatment of a heavily cockled parchment document.  The document was a certificate from the Cinncinnati Society, which honored veterans of the Revolutionary War.  The document was water damaged and had been purposefully cut into several sections, then sewn back together with cotton thread.  There was no evidence of mold and the document had not been lined.

Research into the document’s history revealed that it belonged to a Captain Nathaniel Leonard, who had been suspended from the society for 4 years for ungentlemanly behavior, beginning on July 4th, 1799.  This information supported the hypothesis that the document had been purposefully destroyed and reconstituted, making the sewing important to the document’s history.  The sewing was too fragile to allow the document to be flattened on a vacuum table or by tension drying.  Even flattening the document under pressure put the sewing at risk, as it would not allow the tension to be adjusted as the document relaxed.

Tammy needed a solution that would allow her to both apply tension locally and easily adjust the tension as the document relaxed.  She turned to rare earth magnets for her solution.  Rare earth magnets are available in a variety of strengths and sizes.  Tammy used 11/16” diamater magnets with a profile of 1/32”.  The thin profile reduced the attraction/repulsion between magnets, making their repositioning safe and easy.  The pull force of her magnets was 1.63 lbs, but Tammy wrapped each magnet in a little hollytex bundle to reduce friction, reduce pull force, and a create a handy dandy handle.

The magnets only work, of course, because the document is flattened on a metal surface. Tammy used the following layers, from top to bottom, to protect her work: Polyester film, dry blotter, object, dry blotter, polyester film, dry blotter, metal tray. In the localized areas where Tammy was humidifying the document, she used the following layers, from top to bottom: Polyester film, damp blotter, dry blotter or Gore-Tex, object, dry blotter, damp blotter, polyester film, dry blotter, metal tray. The extra layers between the object and the metal tray help further reduce the pull force of the magnets.

Because of the complexity of the cockling, Tammy realized that she would need to diagram the fiber bundles in the parchment to better understand how humidification would guide the flattening.  Once she better understood how the document would relax, she began working from the inside of the document – applying local humidity – and worked her way outwards to flatten the full document.  The magnets allowed her to see almost all of the document, and she was able to adjust them according to the easily visible tension shifts in the parchment.

Once the document was flattened, Tammy created infills for areas of loss with cast paper and a 3% gelatin solution.  The treated document was string mounted to mat board.  Tammy took special care to attach the string mounts to create extra support around the stitched areas.  The mounted vellum certificate was framed and sealed.

Questions?  Just email Tammy at tamaralynnjordan at yahoo dot com

Doris St-Jacques and Maria Bedynski of Library and Archives Canada presented the challenges they faced when working on the restoration of a hand colored 1739 ed.  L’Amerique Divisee selon Letendue de ses Principales Parties wall map.

Highlights of the talk included images of anthropomorphized, bipedal, dam building beavers (Follow the link for more information on the map itself, and especially the beavers!) and the description of the Library and Archives Canada’s “light wall”.  I don’t think I was the only attendee green with equipment envy at the thought!  I know next time I get to design a new lab space, I will be investing in frosted glass and fluorescent light fixtures.

Library and Archives Canada has two copies of L’Amerique, a map that is important not only for the textual information and hand colored map, but also for the various illustrations of daily life in the Americas.  Their 1689 ed. is in relatively good condition except for heavy trimming, but the 1739 ed. had been brought to conservation in poor shape.  The 1739 ed. had had been previously, and creatively, restored.

• The map had originally been printed on several panels that were glued together in overlapping segments.  The left text panels had been cut from the map during the previous restoration.  As a result, there were definitive breaks between panels that were tenting up and fragmenting at the edges.
• The map had been lined with acidic machine pulp paper (date unknown).
• The map had been trimmed and been given a decorative machine marbled paper border (date unknown).
• The text panels were extensively damaged, and had been in-filled in a variety of creative and baffling ways.
  • Some segments were infilled with replacement printed text.  This text was not only not from the same map, but was not even in the same language.
  • Some segments were decoratively infilled with meaningless scribbles, no doubt meant to emulate text.
• Small fragments from the text had been retained, but had been re-adhereed to the map in seemingly random places.
• The two bottom text panels had been reversed.
• An opaque grey paint had been used to mask areas of loss/damage – further obscuring full lines of text in some areas.

The conservators approached the complex dilemma of L’Amerique by consulting the 1698 ed. of the map which also resided in their collection.  Unfortunately, a previous treatment to that map had heavily trimmed the edges, resulting in significant text loss that made the map unsuitable as a guide.  The conservators were able to purchase a facsimile copy of the map, printed at 1:1 aspect ratio, from the University of Michigan @ Ann Arbor.  Notes on the full treatment of the map are as follows:

The Reboot:

• The full map was humidified, the marbled paper border removed, and the map separated into individual panels.  Humidification allowed the panels to be delaminated where they overlapped, so that the margins were retained where still present.
• Each text panel was pre-treated with alcohol, then immersed in a water bath to remove the paper lining and heavy adhesive layer.
• The two hand-colored map panels could not be immersed in water due to their soluable colorants, so instead, the adhesive and lining paper were removed by laponite poultice.  They were then cleaned by spray misting and light sponging through pasting tissue.
• Gelatin was brushed through pasting tissue to resize the paper of all the panels
• The “creative” infills were carefully removed.

The Repair:

• Leaf casting and hand pulp infills were used to restore areas of loss in each map segment (and to replace the severed margins on the left text panel), and each segment was lined with thin Japanese tissue and allowed to dry.
• The individual sections were then reassembled into the 4 original panels (map, left text, right text, bottom text).  It is at this stage that the conservators were able to switch the left and right sections of the bottom text panel back into their original configuration.

The Reinforcement:

• Terylene fabric was spray-wetted then pasted out with wheat starch adhesive and adhered to the lab’s light wall.  Sheets of kizukishi paper were water torn at the edges and pasted out.  They were adhered to the terelyene to form one large sheet of lining tissue.  The tissue/fabric laminate was then allowed to partially dry to reduce the likelihood that the lining tissue would be disturbed/damaged during the map mounting phase.
• The laminate was repasted out and the map sections (1st the map, then the text panels) were humidified and rolled out onto the kizukishi tissue.  Since the conservators were working on the drool worthy light wall, they were able to easily reach different parts of the map for exact repositioning.
• It was decided to retain the marbled paper because it obscured nothing, would be easy to remove if necessary, and was a part of the history of the piece.  The border was readhered to the map in the original configuration.
• The map was allowed to dry on the light wall.
• Once dried, the map was carefully peeled free from the light wall, with the terelyne serving as a release layer.  The map was placed face down, and the terelyne carefully peeled free from the tissue lining.
• A second layer of water torn tissue was pasted onto the map as a final layer of strength.

The Record:

• From the start, the conservators were concerned about dimensional change during the humidification and lining of the map.  To guard against misaligned fragments, the map was extensively measured and photo-documented before treatment began.
• As the various creative infills were removed,  they were adhered to a mylar overlay in their home positions.
• During the treatment, it was realized that the title banner was delaminating.  Viewed with tramsitted light, it became obvious that the L’AMERIQUE title banner had been pasted over a printed EUROPE banner, and in fact some of the original letters of the EUROPE banner had been altered as a part of  L’AMERIQUE.  The conservators found a that this was also the case on their 1698 ed. map, and that digital images of the same map at other institutions showed a delaminating title banner as well.  So the conservation treatment of this map actually revealed a new fact about the printing methodology used by the map’s creator, Nicolas De Fer.
• A full and complete version of the map text was printed onto a second mylar overlay to be stored with the original map.
• Fragments of the acidic paper lining were retained to save the impression of the original cloth lining still present in the thick adhesive.