Summer 2001 , Volume 7, No. 3



For the past two and a half years, I have been conserving George Washington's last will and testament. This profoundly moving document, which can be considered the General's final message to citizens of the United States, was written in iron gall ink on both sides of twenty-two sheets of stationary. During the Civil War it suffered damage that left all the sheets riddled with tears and split completely from top to bottom. It was masterfully restored in 1910, but ninety years later conservation was necessary again.

The entire, complex treatment will be described elsewhere, but paper conservators may be interested to know how useful fine handmade tengucho tissue proved to be for securely reattaching the split sides and for mending a myriad of tiny tears without obscuring the closely written, double-sided text. Tengujo was also used to mount sheets into double-sided folio mats since Washington's writing extends to the very edges of both sides of the sheets.

Mending Splits and Tears

Almost all mends were made with strips of handmade Tosa tengujo tissue (Hiromi Paper "HM-1", 3.7 grams/sheet, 100% kozo fibers, pH 7.2), as narrow as deemed safe. The width of the mending strips varied with the shape and weakness of torn areas, but generally was about 2mm wide. The length of the strips also varied, from 2 to 25mm. In a very few areas where the writing was pale due to previous damage, thinner (0.001" thick) machinemade tengujo was used.

Sheets of the tissue were toned in baths of Liquitex acrylic paints (artist grade) to various shades of cream and tan so they would be as invisible as possible over the slightly different colors of the will papers. Although acrylic paint becomes insoluble in water once it has dried, the toned tissues were rinsed and redried to insure that any unbound pigment particles were removed and therefore could not transfer to the manuscript. Along the length of the page are splits and very long tears, the color of the tissue was varied, both to accommodate changing paper tones and to diminish visibility of long mends.

It was easy to lift and manipulate tiny pieces of this tissue even when they were wet with paste. I attribute this to three factors: the high quality of the tissue, slight toughening by the acrylic paint binder, and cut, not "water torn," mending strips. The mending adhesive was strong wheat starch paste, diluted with water and then cut with methyl cellulose so the adhesive would not be too strong ("Jin Shofu" precipitated wheat starch and Dow Chemical "Methocel A4M"). The proportion was approximately 1:1 undiluted starch paste and cellulose ether gel.

Mending the center splits on both sides proved sufficient to restrengthen the sheets: facing was unnecessary. I debated whether to mend the other tears on both sides to protect them from snagging and reinforce them better. (Some were long, and the mounting system may strain the sheets more than hinging would.) On the other hand, mending just on one side would maximize immediacy of the object surface and save time and therefore lessen the cost of the project. I decided to mend on both sides except for the smallest tears and losses. Again, mending strips were cut as narrow as deemed safe, generally about 1 to 2mm wide. Cut marks and especially weak fold lines were mended like tears. Thinned areas were reinforced using the thinner machine-made tissue. A few areas of degraded ink were reinforced, using the same tissue and paste.


The fills made for the 1910 restoration were retained because they are made of high quality paper, they fit the losses well, their color remains acceptable in relation to the object, and the 1910 treatment itself is of historic interest. Originally I intended to fill the other, small losses with paper pulp. However, it became evident that slightly more of the writing could remain visible with gossamer tissue supports over these areas so the same toned tengujo was adhered over both sides of these areas.

Tengujo may not be the first tissue one would consider for mending splits and tears, but for this particular object it answered the need to mend securely without obscuring the underlying ink or creating a web of mend lines.

[return to newsletter archive]