Leaf Casting on the Suction Table
by Robert Futernick
Damage to paper takes many forms. The best repair requires use of
a technique that is appropriate to the particular qualities of a
paper, including consideration of intended use of the item. For
example, a hole in a book page could be filled with strong, flexible
Japanese paper to accommodate handling. The same hole in a master
drawing would have different repair requirements, since minimizing
the noticeability of damage would be of greater concern in this
case. The availability of a growing number of repair technologies
enables the conservator to meet the needs presented by particular
problems. Some repair methods are as old as papermaking itself,
while other techniques have evolved during the last 30 years.
Modification of the "leaf casting" technique for use on the vacuum
suction table offers a new refinement for repair and is the subject
of this paper.
Historically, filling holes with similar paper has been the most
frequently used repair method. It can work well, but success will
depend on adequate time to do the task, a good selection of repair
papers, a minimum of edge discoloration, and the skill of the
conservator. Problems can develop, however. Secure joining of the
insert to the original may be difficult, especially with thick,
brittle paper. Also, distortion of the original sheet can develop
because of the different expansion characteristics of the two papers
(new and old) or the introduction of adhesive necessary for joining
Because of its strength and flexibility, Japanese paper has been
widely used for book leaf repair. Two or more thin sheets laminated
together and slightly overlapping the original can create an
excellent fill. Though different in color and texture from many
Western papers, Japanese paper inserts often provide a sympathetic
match. This method seems to have less propensity for paper
distortion than other techniques. Furthermore, the strain of
attachment is spread over a larger area. This may be extremely
important when the edge of the original sheet is weak or the
intended use of the material involves extensive handling.
In the late 1950s, Esther Alkalay and Yulia Petrovna Nyuksha
began experimenting with the concept of filling missing areas in
paper with fibers suspended in a liquid medium. By 1961, equipment
for this purpose had been developed and was in use in laboratories
located in Eastern Europe and the U.S.S.R. The process has come to
be known as "leaf casting", and is best described in the preprints
of the 1980 Cambridge Conference (see bibliography). Today the
Library of Congress and the Northeast Document Conservation Center
both use leaf casting machines based on Alkalay's design. Other
models, varying in cost and sophistication, have been developed in
recent years and are in use in laboratories throughout the world.
Steps of Traditional Leaf Casting:
- The damaged paper should be thoroughly wet. (Extending the soak
time and increasing the wetness of the sheet aids in later bonding.)
- The paper is placed on a permeable surface at the bottom of the
leaf casting tub. A hold-down mechanism is usually employed to
restrict movement of the damaged sheet when water is added to the
area above the paper.
- A precise amount of pulp of appropriate fiber type and color is
stirred into the water bath. It is important that measurement of the
missing area and paper thickness be accurate and the pulp addition
be correct so that the cast fills will equal the thickness of the
original paper. Simple graph paper or a planimeter can facilitate
- Suction, whether from an electric pump or gravity, is applied to
the space below the damaged work. Water in the top portion drains
through the voided spaces only, depositing an even layer of pulp
across the loss.
Characteristics and Qualities of Leaf Casting
Leaf casting has distinct advantages over other methods of
filling losses. If one's work is organized and equipment is properly
adjusted, damage to paper can be repaired very quickly. It is quite
possible to create a smooth, flexible and unobtrusive transition
from the fill to the original paper. And if fibers of similar
dimensional characteristics are selected and prepared, distortion
often imposed by a fill can be minimized or avoided. However, there
are areas of concern and potential problems in leaf casting, and
they should be understood in order to take full advantage of this
The bonding of the fill material to the old paper can be
problematic. The quality of connection will depend on a) the state
of deterioration of the original sheet; b) the bonding
characteristic of the pulp prepared for casting; c) the edge of the
loss and the degree of fiber extension; and d) the speed that water
drains through the losses.
In the past, various kinds of adhesives have been added to the
casting liquid to improve bonding. Unfortunately, adhesives affect
the entire object during the period of immersion and drainage, not
just the fill. This changes the nature of the paper—at the
very least—and if the adhesive material proves to have poor
aging characteristics, the entire artifact will suffer. Therefore,
the addition of adhesives, though helpful in bonding, should be
considered and avoided when possible.
Conventional leaf casting requires the complete immersion of the
paper in water. Clearly, many items cannot undergo this action
without risking loss or alteration of the media. As a remedy,
application of fixing agents has been suggested. However, it seems
imprudent to impose such an altering measure just for the purpose of
repair when other satisfactory methods are available. Hand pulp
technique and localized leaf casting methods do not require total
During leaf casting, there is some pulp deposition onto the
surface of the original paper. As the fluid level in the tank gets
low during the final stage of drainage, some of the fibers in the
solution will become affixed to the surface of the original. This
may be minimized if the pulp-in-fluid concentration is kept low,
though in certain cases some cleanup may be necessary. (If the work
is one-sided and the casting takes place with the work face down,
then slight deposition on the verso may not be an issue.)
In an early publication too obscure to remember, reference is
made to the employment of chronic gum chewers for major restoration
ventures. Here, beating and fibrilation of fibers with cuspid and
molar action is gentle but thorough, with chemical processing
provided by enzymatic mechanisms (spit). Unfortunately, with leaf
casting activity on the rise, it has become more and more difficult
for conservators to locate eager chompers to fill increasing
Many conservators have turned to the use of household blenders
for conversion of old paper to pulp. Recently, blender preparation
has been criticized in the literature in favor of beating devices
especially designed for the purpose. Certainly, beating machines do
provide precise control over fiber length and hydration and they
will accommodate the use of raw fibers. Petherbridge points out that
bonding and dimensional stability is very dependent on pulp
preparation. However, the use of expensive equipment is not always
practical or necessary. If care is shown in the selection of paper
for repulping and the fiber-to-water ratio is properly controlled
during beating, the quality of blender-beaten fills may be quite
acceptable for many purposes.
Hand Pulp Technique
The development of hand pulp technique was concurrent with the
invention of mechanical leaf casting. At the AIC Conference in
Dearborn in 1976, Keiko Keyes outlined the technique of hand pulp
casting that she developed and perfected. Keiko demonstrated how a
hand method, depending on the skill of the conservator, could be
successfully used to repair damaged artwork quickly without
requiring expensive equipment. Because the entire sheet does not
require complete wetting, artifacts with media too vulnerable for
immersion casting can be safely treated with hand technique. Also,
more control may be achieved with hand placement of fibers than with
mechanical leaf casting. Damaged paper with small-to-medium-sized
holes are perfect candidates for this method, though larger losses
in smooth regular papers are difficult to fill evenly by hand. A
combination of hand and mechanical methods for use on a vacuum
suction table provides a practical alternative to such problems.
When Marilyn Weidner introduced the vacuum suction table at the
1974 AIC meeting, it was difficult to imagine all of its possible
applications. Since that time, it has become such an important tool
in most laboratories that paper conservators often wonder how they
could practice without it. Many people have built, modified,
analyzed and used suction tables over the last eight years.
Inexpensive home-made models can serve quite well, as can the more
expensive and elaborate commercially available models. Important
factors to consider for both general use of the vacuum table and
leaf casting are:
- Openness of surface. Thick masonry tops tend to restrict air
flow and become clogged.
- Flatness of surface. Any material other than very fine screen or
some perforated metals will impose their texture on the paper or
blotter in contact.
- Suction. The pump/fan arrangement should have the capability of
lifting a column of water at least 80 inches high.
- Air flow. This will vary depending on the size of the table.
Generally 20 cubic feet per minute (CFM) per square foot of table
surface is a reasonable guideline. A table 2' x 3' (6 x 20 CFM)
would perform quite well for most purposes with a suction system
that could deliver 120 CFM.
Leaf Casting on the Suction Table
Until recently, the suction table has been primarily used for
washing, drying, stain removal, and lining operations. With only
slight modification, the suction table can also be used for
localized leaf casting. While hand pulp method remains the simplest
and most direct approach for small holes, leaf casting on the
suction table may be especially useful under certain conditions:
- In the absence of a conventional leaf casting apparatus.
- When the art object or document is larger than the dimensions of
the leaf casting machine.
- When the loss is large and requires a particularly even fill.
- When the texture of the original sheet is an aesthetically
important element and the repair calls for articulation of paper
Advantages of suction table casting:
- The entire sheet does not always have to be wet before casting.
- When water-sensitive media is present, another liquid such as
alcohol may be utilized to carry the pulp into place.
- Pulp deposition on the original paper is avoided, since very
little liquid is used and only a small area of the original is
- The quality of the join can be improved in cases where strength
is important by casting half the fill and then turning the sheet
over to complete the fill. This creates an actual encasement or
overlapping of the edge on both sides. This variation is not only
strong but offers the additional advantage of obscuring a dirty
Modification of Suction Table for Leaf Casting
With only slight modification, many vacuum/suction table designs
may be used for leaf casting. A major issue is the ability of the
system to safely handle a gallon or more of water at one time. The
liquid should be disposed of before it reaches the suction device.
This can be accomplished in several ways:
- Installation of a tee in the tubing leading to the pump/fan. The
pipe fitting should be placed so that liquid will fall into a trap
for collection and future disposal (see figure
2). Plastic (PVC) tubing and fittings are especially useful for
the plumbing. The pipe can be easily cut to size with a handsaw and
simply slip-fitted for connection. Though joins can be welded with
solvent for permanent high pressure usage, this is not recommended.
It is an advantage to undo a connection for later adjustment or
- Water collection in the plenum space below the table surface. If
this space is large enough and waterproof, it can serve as a trap.
Vacuum input must be placed high with a removable plug positioned at
a low point for water removal (see figure 3).
Leaf casting on the suction table does not differ in principle
from the conventional method except that the casting fluid is
contained only in the area directly above the lacuna. A discussion
of the process follows:
- The pulp container (see figure 4). Round
plastic tubing can be procured in various sizes (2" to 6" in
diameter) and easily cut into 4" lengths .
(Containers for larger losses are easily fashioned by heating and
bending plexiglas sheeting into a rectangular shape, slightly larger
than the hole). Thin plastic film (.002") is cut into a rectangular
shape several inches larger than the diameter of the tubing . Then a hole, larger than the loss but slightly
smaller than the opening of the container tube, is cut in the center
of the rectangle . Silicon adhesive applied
to the bottom edge of the tube affixes the container to the plastic
mask . This pulp mold will seal itself when
suction is applied without harmful downward pressure of the mold on
the paper artifact.
- Arrangement of materials on the suction table (see figure 5). A piece of thin plastic film cut larger
than the top of the suction table is placed on the surface so that
the extending edges hang-over the table . A
hole in this film is cut centrally, its dimension slightly larger
than the opening of the container mask. Two layers of non-woven
spunbonded polyester fabric (Reemay, number 2014 or 2114 from
Dupont) are placed over the opening to provide a surface on which to
cast . Reemay, an especially suitable
material, aids in creating very even pulp formation and imparts a
texture much like that achieved by contact with papermaker's felts.
The damaged original is set on the Reemay so that the lacuna is
located over the opening in the table . The
container can be set directly on the surface of the paper . However, the opening in the mask will not
conform exactly to the contour of the loss. If precision in casting
is required, pulp deposition in areas other than the hole can be
avoided by cutting and properly placing a Reemay mask onto the paper
. This extra step can be done quickly and
enables one to use just a few different-sized containers for most
- Delaying drainage. An interval of time is needed for filling the
container with water and pulp. A removable plug can be fashioned
using plastic film, cut so that it just fits in the container tube
(see figure 5 ). A
string or narrow strip of plastic affixed to one edge of the plug
facilitates removal. The plug, inserted inside the container just
prior to applying suction, is pulled tightly against the container
mask forming a temporary bottom to the container when suction is
activated. Lifting the attached string removes the plug, allowing
the slurry to be pulled into place. Larger areas of loss are plugged
better if several pieces of plastic film (each with a tail) are
placed to overlap each other slightly. Multiple plug removal becomes
a louver action, thus avoiding tidal waves.
- Correct addition of pulp to casting liquid. Too much pulp yields
a fat, irregular, uneven casting and too little will produce
transparency and thinness. (As previously mentioned, it may be
desirable in some cases to cast a thin fill from one side, turn the
paper over, and complete the casting from the other side. This
encases the edge of the original, providing more strength at the
join, while perhaps hiding a dirty edge.) The achievement of a fill
that has comparable thickness to the original as well as appropriate
texture and opacity is a somewhat complicated matter. Practicing on
holes in mock-ups is necessary to master this technique. However, it
is helpful to have a starting point and the following indicates a
- Cut a rectangle of paper of the kind that will be used for
pulping to a convenient size (e.g., 8" x 10"). Measure and record
the thickness and weight of the sample (e.g., .005" thick and 4.55
- Determine volume-per-gram by multiplying the length of rectangle
times the width times the thickness and divide by the weight of
sample (e.g., 8" x 10" x .005 divided by 4.55 grams=.0879").
- Find the area of the loss and measure the thickness of the
damaged artifact. Counting squares of graph paper or outlining the
perimeter of the hole with a planimeter are two methods that have
been used to approximate area. Either technique is adequate for
occasional casting. However, if the volume of leaf casting activity
is high, it may be worth speeding area calculation with the aid of a
microcomputer. Simply tracing the outline of the loss with a "light
pen" on a Graphics Tablet connected to an Apple Computer, for
example, will yield a very quick and accurate area measurement.
- Determine the weight of pulp required for the casting by
dividing the volume of the loss by the volume of a gram of pulp
(e.g., area of loss 1.875" x thickness .007" XXX/ .0879" =.1493
- It is convenient to store pulp in solutions of known
concentration (e.g., 1%). It is then easy to mix different colored
pulps, test the resulting color by drying, and then pour out the
required amount (e.g., .1493 grams is approximately equal to 15 ml
of a 1% solution; more accuracy is achieved if solutions are kept in
even lower concentrations).
- With suction holding container and plug in position, the pulp
solution is poured into the container. Vacuum pressure is then
adjusted. This is a matter of experience, but pressure influences
flow and the flow rate will affect the casting. The tail of the plug
is quickly pulled. This action opens the bottom of the container
enabling the liquid to drain.
- The casting is quite wet and much water can be removed by
slipping a sheet of blotting paper between the first and second
layers of Reemay. Acting as a support, the Reemay can then be lifted
to move the paper to a hard-surfaced table. Rubbing over the fill
with a bone burnisher while it is still wet serves to compact fibers
and improve the join. Using plastic film or polyester fabric as an
interleaf during burnishing will help to achieve a more sympathetic
match of textural qualities. Subsequent drying of the sheet in open
air is often preferable to pressing even under light weight. Large
castings have a tendency to detach at the join and this can be
avoided by air drying. Re-moistening for final pressing and
flattening is accomplished with controlled, even water application.
By this time, the joined areas are more secure and not as
susceptible to splitting.
- If the addition of adhesives to the fill seems
appropriate—for strength or as a sizing for later
toning—then it might either be added to the casting solution
or applied locally after the fill has undergone preliminary air
- In special cases, it is desirable to approximate the textural
qualities of the original in the casting. Articulation of laid lines
in the cast area can minimize the visibility of damage. This becomes
important with a work of art on textured paper scheduled for
exhibition in subdued and slightly raking light. The use of a laid
screen, aligned on a light table to match the laid markings of the
paper and placed just under the loss during the casting process, can
impart a sympathetic texture to the fill. A method for creating an
assortment of laid screens for this purpose is described as follows:
- Make a ruled line drawing using black ink on clear acetate. The
width of the lines should be drawn to equal the space in between the
lines. Lay out the grid so that approximately 24 black lines are set
to the inch (see figure 6).
- Using a photostat machine, incrementally enlarge and reduce the
line drawing to achieve a desired range of sizes.
- Place a line drawing in direct contact with a piece of
photo-silk screen emulsion. The photo-sensitive emulsion is a
commercially available product mounted on a plastic film support.
One company that supplies this material is J. Ulano & Company,
610 Dean St., Brooklyn, N.Y. 11238.
- Set a sheet of glass on top of the sandwich and expose the
emulsion to light through the open areas of the line drawing. A 250
watt sun lamp is a good source and it will work well if placed 15-20
inches from the drawing. Exposure will probably be under 12 minutes
but a little testing is required to achieve the proper duration
- After exposure, place the emulsion into a liquid developer
specific for this purpose. Rinse according to the emulsion
directions and while the emulsion is wet and sticky, press it onto a
support of Reemay or fine (polyester) silk-screen fabric. Allow to
dry and then remove the plastic film carrier.
- The dried emulsion remains sensitive to water and must be
hardened before use. Expose the silk screen to the fumes generated
by a formaldehyde bath and allow to dry. Then spray the screen with
a material such as Scotchguard to further protect the emulsion.
- Repeat this process for each line drawing to create a catalog of
various-sized screens that can be used again and again.
- It should be noted that this process is not for the creation of
a perfect reproduction or a facsimile of a missing area. Instead,
its use is intended to minimize the visibility of damage that would
otherwise interfere with the aesthetic impact of a work of art.
- Casting with alcohol. There are instances where even partial
wetting of a paper artifact is not safe. Experiments have shown that
an acceptable casting can still be accomplished, providing the paper
or media is not adversely affected by alcohol. The pulp is prepared
by first beating it in water in the usual manner. Most of the water
used in beating is extracted by pouring the slurry through a
strainer. The damp fibers are transferred back to the blender where
a quantity of ethanol is added for a second beating. The paper
artifact is wet with ethanol prior to placement on the table and
casting is performed as previously described. The resulting fill
will be quite soft and felt-like because not much bonding occurs.
However, this may be remedied by localized application of water or
water plus a sizing agent followed by burnishing and pressing.
Alteration of the Vacuum Table Design
Recently we constructed a small portable vacuum table that
especially lends itself to leaf casting and pulp filling procedures.
It is made entirely of clear plastic, so that when it is placed on a
light table, light is transmitted up through the top of the suction
box. Permeability of the upper surface was achieved by drilling many
small holes in the plexiglass. However, Marjorie Cohn suggested at
the Milwaukee meeting that a fritted glass plate might be set in the
top surface to allow transmission of light without the need for
drilling. This excellent idea would make construction quite easy.
Working features of a transmitting suction table:
- A light suction table is extremely versatile. It can serve for
leaf casting as just described, conventional pulp filling, and for
very precise and selective addition and deposition of pulp.
- For conventional pulp filling, suction is applied but
temporarily blocked by placement of a sheet of thin plastic film
over the opening of the table. When proper pulp density and
distribution is achieved, the "table cloth" (plastic film) is yanked
and the liquid quickly falls, leaving the fibers in place.
- For certain shaped holes or art that cannot tolerate much
wetting, a transmitting table is extremely useful. A dirty edge or a
small loss, tear or paper thinning is set on Reemay over the suction
area with the vacuum on. Pulp, suspended in very dilute
concentration, is dripped onto the damaged area with a wide tipped
medicine dropper. Air flow takes the pulp to the open area and
causes adhesion to the edge of the top or to the previously dripped
application. It is possible to build up multiple layers until the
desired result is achieved without much lateral wetting. It is
necessary to send distinct droplets. Attempting pulp application
from close range will result in fiber clumping.
The leaf casting technique offers a quick and viable alternative
to traditional repair techniques. Conducting this kind of procedure
on a common suction table increases both the number of artifacts
that can be treated and the number of laboratories capable of doing
the work. The use of transparent suction tables and silk-screens
designed to impart texture to the castings provide further
refinement now available to the conservator.
The success of this project was dependent on the help and
patience of all the people in our laboratory, namely Pauline Mohr,
Patricia Morris, Linda Ogden, Sylvia Rodgers, and Janice Mae
Schopfer. Special acknowledgement goes to Keiko Keyes. It was
through several meetings with Keiko that the original concept was
refined and simplified. Finally, this paper became much more
readable after the careful attention of Jennifer Futernick. She had
just cut her editorial teeth on a major book (In Search Of
Excellence... published by Harper and Row) so this paper was
mere baby food.
1. Alkalaj, Stella: "The Chemical Laboratory for Hygiene,
Conservation and Restoration of Damaged Written Materials,"
Restaurator 1:2 (1969), pp. 87-91 (Copenhagen,
2. Wächter, Otto: "Methods of Restoring Old Prints,
Documents, and Drawings Using Liquid Paper Pulp," Lisbon Conference
1972, Conservation of Paintings and the Graphics Arts,
pp. 971-974. (IIC publication, 1972).
3. Alkalay, Esther Boyd: "The History and Development of Leaf
Casting," The Conservation of Library and Archive Materials
and the Graphics Arts, Cambridge 1980 (Preprints), p.187.
4. Petherbridge, Guy: "Analysis, Specification, and Calculation
in the Preparation of Leaf Casting Pulp: A Methodology," The
Conservation of Library and Archive Material and the Graphic
Arts, Cambridge 1980 (Preprints), pp. 189-209.
5. Keyes, Keiko Mizushima, and Farnsworth, Donald S., "Practical
Application of Paper Pulp in the Conservation of Works of Art on
Paper," pp. 76-86, Preprints of papers presented at the fourth
annual meeting of the American Institute for Conservation,
6. Perkinson, Roy, and Futernick, Robert, "Questions Concerning
the Design of Paper Pulp for Repairing Art on Paper,"
Preservation of Paper and Textiles of Historic and Artistic
Value, No. 164, 1977 (John William, editor; published by
American Chemical Society).
7. Weidner, Marilyn Kemp, "A Vacuum Table of Use in Paper
Conservation," Bulletin of the American Institute for
Conservation Vol. 14, No. 2 (1974).
8. Du Meer, Robert, "Construction of a Leaf Caster for Small
Laboratories," Preprints of paper presented at the sixth
annual meeting of the American Institute for Conservation,
Elements of this tale may
have basis in fantasy.
Received: Fall 1982
Paper delivered at the Book and Paper specialty group session, AIC
10th Annual Meeting, May 26-30. 1982, Mailwaukee,
Papers for the specialty group session are selected by committee,
based on abstracts and there has been no further peer review. Papers
are received by the compiler in the Fall following the meeting and
the author is welcome to make revisions, minor or major.