Conservator Edward Cheese has recently completed the repair of a medieval pocket Bible (B.10.21) in a project which, owing to its complexity and the Covid lockdown, took over three years to complete. In this blog he describes the meticulous conservation process.
Prior to conservation the manuscript was in an extremely fragile state: the opening leaves were disintegrating and the inflexibility of the spine made opening the book almost impossible.
Historically the book had been damp for a prolonged period. Mould had grown extensively in the parchment of the opening quires and the iron-gall ink had eaten through the thin membranes in the worst-affected areas. The binding in which the book had come down to us was put on in the 18th century apparently in an attempt at repair. Unfortunately, the large amount of hide glue applied to the spine caused the already degraded parchment to become gelatinised. Many of the most fragile leaves had split away from the binding along the line of the glue penetration and the large majority of the spine folds were hard and brittle.
In order to consolidate the extremely fragile leaves of this manuscript, it was eventually decided with the Librarian that the book should be disbound. This also allowed for documentation and assessment of the complex collation of the book.
Loose fragments of parchment were saved before the boards, endleaves, endbands and spine-covering were removed. This revealed the thick, hard layer of hide glue on the spine which was softened and scraped away a little at a time. As the final deposits were removed, the sewing threads were cut and the quires, where they existed with intact spine folds, were eased apart using a very fine bone folder and an extremely cautious touch. Groups of single leaves were separated later, as the spine edges in these cases were a hard, gelatinised amalgam of degraded parchment fragments and glue. In most cases, the fragments were already splitting from the un-glued parchment.
It was immediately obvious that the highly degraded parchment was extremely sensitive to moisture: distorted areas could be gently straightened out using the moisture in the breath, directed with a paper tube, rather than by humidifying equipment, and tests on small fragments showed that even the most controlled applications of adhesive for repairs/consolidation of the leaves caused staining and distortion. Gelatine and paste were therefore rejected as adhesives for the consolidation part of the project, and a fine remoistenable tissue was selected to control moisture content.
The parchment leaves were consolidated and repaired with 2.5-3.5mm dots of remoistenable tissue cut with a Japanese screw punch and applied with very fine tweezers under magnification. Where it was safe to do so, each leaf was cleaned of surface dirt and dust using a very soft brush before repairs were applied. Damaged spine folds were supported with shaped patches and guards of the remoistenable tissue adhered with a viscous solution of methyl cellulose in purified water, to which methylated spirit was added to further reduce the water content and accelerate drying.
The treatment of the parchment, although laborious and extremely time-consuming, was a success. However, the repairs were designed to consolidate the very fragile leaves, not make them robust enough for binding and direct handling. It was therefore decided that, rather than attempt to rebind the manuscript as a book, each quire should be reconstructed incorporating interleaving of archival paper cut larger than the original leaves. The interleaving would allow the leaves to be turned without direct handling and prevent face-to-face contact of leaves.
Each interleaved quire was then sewn as a pamphlet with a paper cover, and the whole set of quires and the binding fragments were housed in a bespoke drop-spine box with hinged covers designed to keep the quires under the slight pressure when the box is closed.
Due to the fragility of the original leaves, access to the manuscript is extremely limited. The conserved manuscript has been photographed and can be read in the Wren Digital Library. Those interested in the collation determined during the disbinding can view the diagrams.