Applied Decoration for Historic Interiors: Preserving Composition Ornament
Anyone who has ever walked through historic houses and large publicbuildings, visited an art gallery, picked up a picture frame inan antique shop, or even ridden on an old carousel has been closeto composition ornament, but has probably not known whatit was or how it was made. This is not surprising, since compositionor "compo" was conceived as a substitute for more laboriouslyproduced ornamental plaster and carved wood and stone, so wasintended to fool the eye of the viewer (see Fig. 1). The confusionhas been heightened over time by makers who claimed to be thesole possessors of secret recipes and by the variety of namesand misnomers associated with the material, including plaster,French stucco, and Swedish putty, to name a few.
Many natural or man-made materials can be made soft or "plastic"by the application of heat and are called "thermoplastics."Composition is a thermoplastic material used to create sculpturalrelief. It is soft and pliable when pressed into molds; becomesfirm and flexible as it cools; and is hard and rigid when fullydry. Typically formulated with chalk, resins, glue, and linseedoil, this combination of materials gives compo its familiar light-to-darkbrown color. It is the only one of the so-called thermoplasticmaterials to be used extensively in architectural decoration becauseof its low cost.
Generally adhered to wood, historic composition ornament is mostoften found decorating flat surfaces such as interior corniceand chair rail moldings, door and window surrounds, mantelpieces,wainscot paneling, and staircases-indeed, anywhere that buildingdesigners and owners wanted to delight and impress the visitor,but stay within a budget. While composition was cheaper than carvedornament, it was still meticulously hand made and applied; thus,it was more often used in "high style" interiors. Butthe types of structures historically decorated with compositionornament were more democratic, encompassing residential, commercial,and institutional buildings, and even including specialty applicationssuch as the social saloon of a steamship (see Fig. 2).
With proper understanding of the material, historic compositionornament may be successfully cleaned, repaired, or replaced insections. Unfortunately, because composition is often misidentifiedas plaster, stucco, or carved wood, the use of inappropriate methodsfor removing paint is a major cause of its loss (see Fig. 3).The purpose of this Brief is to to assist historic property owners, managers, architects, craftsmen,and preservationistsin identifying existing composition ornament, determining theextent of repair and replacement needed and, finally, selectingthe most sensitive, non-destructive method of treating it.
While various types of moldable composition date to the ItalianRenaissance, architectural use of composition did not begin toflourish until the last quarter of the 18th century. During thisperiod, many composition ornament makers in Europe and Americasupplied the public with complex sculptural decoration. Also,the overly complicated and often intentionally mysterious earlierrecipes were now reported to be comprised of a few basic ingredients:animal glue, oil (usually linseed), a hard resin (pine rosin orpitch was cheapest), and a bulking or filling material, generallypowdered chalk or whiting (see also Sidebar, Compo: TheBasic Ingredients).
Compo: The Basic Ingredients (clockwise from front center)(sidebar and photo)
Chalk: Chalk is whiting in solid form. It is a typeof white, soft limestone.
Glue: Before the invention of synthetic adhesives,glue meant animal or hide glue. This was made by boiling animalskins to extract a protein-collagen-in water, then condensingand drying the collagen until it was in solid form. A varietyof types and grades were, and are still, available. Two are shownhere.
Linseed oil: This is a yellowish drying oil obtainedfrom flaxseed that is used in paint, varnish, printing ink, andlinoleum; it is a key ingredient in composition ornament.
Resin: Resins are organic materials present in woodand exuded from various trees and shrubs. In unrefined form, theyoften consist of a mixture of solid natural polymers, oils, andvolatile aromatic substances.
Compo mixes have been the subject of a good deal of variationand there has never been a set recipe, but the ornament manufacturersof the later 18th and early 19th centuries understood in generalterms what their material was and what it could do (see Fig. 4).The advantages of the material were described by a prominent Americanmaker, Robert Wellford, in his advertising broadside of 1801:
"A cheap substitute for wood carving has long been desirablefor some situations, particularly enriched mouldings, etc., andvarious were the attempts to answer the purpose, the last andmost successful is usually termed Composition Ornaments. It isa cement of solid and tenacious materials, which when properlyincorporated and pressed into moulds, receives a fine relievo;in drying it becomes hard as stone, strong, and durable, so asto answer most effectually the general purpose of Wood Carving,and not so liable to chip. This discovery was rudely conductedfor some time, owing to Carvers declining every connection withit, till, from its low price, it encroached so much upon theiremployment, that several embarked in this work, and by their superiortalents, greatly improved it."
In brief, compo is perhaps best understood as an early thermoplasticthat allowed the rapid reproduction of complicated detail forpopular use.
Since the craft has essentially remained the same over time, adescription of its historic manufacture is also applicable today(see Fig. 5).
In one container, chunks of amber colored pine rosin or the cheaperblack pitch were heated in linseed oil until they melted togetherand combined completely. In another container (often a double-boiler),previously soaked chunks of animal glue derived from skins andhides were cooked and blended into a uniformly thick solution.The two liquid components were then stirred together. This "batter"was made into a pliable "dough" in a way familiar toany baker. It was poured into a cratered pile of whiting and firstmixed with a spatula until it was thick enough to be kneaded by hand. Vigorous folding andkneading inof more whiting was done until the composition had a consistencylike modeling clay and was completely uniform.
To mold a decoration, the compo was first warmed in a steamer,and the mold prepared with a thin coating of oil and a dustingwith talcum powder. A piece was then kneaded and folded to producea smooth and wrinkle-free surface on one side. The good side wasplaced down over the rigid mold, and pressed in loosely with thefingers, leaving excess above the surface of the mold. A dampboard was placed over this and the "sandwich" placedin a screw press and squeezed so as to force the compo into thefinest detail. It was then removed from the press and turned overso that the mold could be lifted straight up, leaving the compostuck to the board. Upon cooling to room temperature, the compogelled, becoming tough and rubbery (the gelling property is dueto the glue component which is chemically identical to ediblegelatin). At this stage, it was sliced off the board with a thin-bladedknife. The remaining mass of composition still adhered to theboard could also be sliced off and reused.
Composition ornament was often fixed to an already prepared woodensubstrate at the factory while it was still fresh and flexible,but could be dried and shipped to the final user, who would makeit flexible again by steaming on a cloth stretched over a containerof boiling water. Instructions for doing this, as well as suitablebrads for "fixing," were supplied by some manufacturers.Because of the glue component, steaming the backs of ornamentswould make them soft and sticky enough to self-bond without additionalglue. Soft ornaments were softened nailed through or pressed downon top of previously driven headless brads (also called sprigs).Strings and wires were often included in the mass during pressingto serve as internal armatures and reinforcements. These measurespreserved the integrity of the ornaments even if they cracked.
Originally meant to copy other materials such as wood, plaster,and stone, composition had its own unique properties and advantagesthat were soon exploited in both technical and artistic terms.It has distinct characteristics in each of its three states: pliable,rubbery, and hard. When warm and pliable, it can be modeled bya skilled worker and it is capable of receiving the finest detailwhen squeezed into a mold. After it has chilled to room temperatureand is gelled, it is rubbery, flexible, and tough. The detailis essentially set and cannot be easily damaged as the ornamentsare manipulated (see Fig. 6).
Gelled composition ornaments can be easily bent over curved surfaceswithout cracking, and unlike a rigid cast material such as plaster,they can be stretched or compressed somewhat to fit a design withoutdamaging the detail. An egg and dart motif, for example, couldbe made to come out evenly at the corners without making a partialegg or dart. The sculptural vocabulary from the maker's mold collectioncould be re-arranged at will into larger decorative schemes. Infact, any smaller component of a decoration from a single moldcould be sliced free and inserted into any location.
Composition could be carved to heighten detail, correct defects,or undercut ornaments-that were, of necessity, straight-sided-sothat they would release from the rigid molds. This could
be done in the gelled state or, with more difficulty, after ithad finally hardened to stone-like solidity.
Finally, when completely hard, it could be given a polished marbleshine with nothing but a damp cloth. It could be stained, coatedwith any sort of paint, varnish, or oil gilded without any furtherpreparation (see Fig. 7).
A technical discussion of composition is not complete withoutan examination of the molds used to create the ornament. Thesewere the ornament maker's largest investment in time and expense,and were the key to the craft (see Fig. 8).
Composition molds were always made of rigid materials that wouldwithstand the considerable pressure used in pressing the ornaments.All of these materials and methods have been used in sculpturalcrafts since the Renaissance. The comparative listing that followshelps explain their advantages and disadvantages.
Wood was carved in reverse to create a negative matrix.This was highly skilled work often performed by a specialist carver,and required a large initial investment in time, but wooden moldswould essentially last indefinitely if properly maintained. Afurther design advantage of reverse carving is that fine incisedlines will show up as fine raised lines in the final ornament.(Fine raised lines are notoriously difficult to carve or modelin relief.) Molds carved from dense and close-grained fruit woodssuch as apple and pear seem to have been common in the 18th century.In the 19th century, the most intricate molds were carved in boxwood,often encased or framed by larger and cheaper pieces of timberfor ease of handling and to prevent splitting.
Metal alloys such as brass, bronze, and pewter made excellentmolds capable of yielding the highest level of detail and werevirtually indestructible in use. They were expensive due to theintrinsic value of the metal and because their production involveda variety of complex and skilled steps performed by modelers,pattern makers, and founders. Few historic metal molds have survived,possibly as a result of wartime scrap drives.
Sulfur melts into a clear fluid at about 115 C and couldbe poured over a positive clay model or another compo ornament.A sulfur mold resembles hard plastic, but is more fragile. Evenwhen framed in wood and reinforced with iron fillings, as wascommon practice, it was especially vulnerable to breakage. A figuraldesign, such as a frieze of The Three Graces, wasmuch easier to model in relief than to carve in reverse, and sulfurwas one of the few materials that could be used to make a hardmold from a clay model.
Composition itself could be squeezed over a hard reliefpattern (such as another manufacturer's ornament) to make a mold.Composition shrinks as it hardens and so the mold was always smallerthan the original. It is also fairly brittle when hard and, likesulfur molds, would tend to crack in the press. Composition "squeezemolds" were ideal for pirating another maker's patterns!
Pitch molds became popular during the late 19th and early20th centuries. A warm and soft mixture composed primarily ofpine pitch was poured into a recess in a wood block or frame.It was then turned over and squeezed down onto an oiled woodenpattern. Pitch molds might crack with age or in the press, butas long as the carved pattern was retained, they could be easilyre-made.
Early History and Renaissance. Press-molded decorationhas been used with various soft plastic materials for centuries.For example, it is known that medieval sculptors press-moldedorganic mixtures to decorate painted sculptures. But because mixturesbased on organic binders such as glue, oil, resins, and waxesare prone to various sorts of degradation, actual survivors arerare.
The direct ancestors of the composition craft are most likelyfound in the Italian Renaissance; however, composition mixtureswere not extensively used for architectural decoration duringthis period, probably due to building traditions as well as relativeexpense. It is worth nothing that this was an age of experimentationwith materials and rediscovery of Greek and Roman designs. Pressmolded mixtures called pastiglias were used to decoratewooden boxes and picture frames as early as the 14th century (seeFig. 9). Moldable compositions were discussed by various Renaissancewriters. The recipes are extremely varied and include, among theirmore common and understandable ingredients, gypsum, lead carbonate,wood and marble dust, eggs, pigments, sheep's wool, and variousoils and resins.
The 18th Century. The first flowering of architecturalcomposition in America took place at the end of the 18th centurywhen ornaments were both imported from England and produced bymakers in every major eastern city. All of the conditions wereright: molding technologies were well established (architecturalpapier mache, which, like composition, was produced inmolds, had gained widespread acceptance during the middle decadesof the century). The raw materials were produced or imported involume, so the cost of the composition ingredients came down asthe cost and availability of highly skilled labor went up. Economicand social conditions favored centralized "manufactories"in the production of various arts and crafts.
Design trends also fed into a favorable reception for composition.A more faithful reinterpretation of Greek and Roman design eventuallytermed "Neoclassical" had taken hold in Europe, championedin England by the architect, Robert Adam, after his return fromstudy in Italy in 1758 (see Fig. 10). Although Adam played nodirect role in the "invention" of composition ornament,as has sometimes been said, he patronized English craftsmen whowere making it and was generally receptive to new and innovativematerials. One early maker, sometimes cited as the "inventor"of composition by his contemporaries, was John Jaques. His nameappears in London advertising by 1785, but he was probably inbusiness before then (see Fig. 11).
As a result of Adam's influence, designers of applied ornamentin both Europe and America began to take advantage of a moldingprocess that was ideally suited to producing the detailed, butrepetitive, motifs of classical decoration-acanthus leaf, eggand dart, festoons, swags, and paterae-as well as classical themesdepicting Greek and Roman gods and goddesses (see Fig. 12). Andas the Neoclassical style became more popular, composition ornamentmakers increased in number.
The 19th Century. During the early decades of thenineteenth century, Neoclassical-encompassed in America by theterms Federal, Empire, and Greek Revival-was in the ascendancy.Composition makers continued to increase and also to find newuses for their material. Composition picture and mirror framesbecame common and some makers advertised the suitability of compositionornaments for casting iron firebacks and stoves. Composition ornamentwas explicitly advertised for exterior use as well, although verylittle has survived. The interiors of houses and public buildingsin every prosperous American city were decorated with composition(see Fig. 13).
When the classically derived Federal and Empire styles gave wayto the various revival styles-Rococo, Gothic, Renaissance, andItalianate-composition makers simply made new molds to accommodatethem. (Although Rococo and Renaissance styles were not commonfor architecture in America, they were common for furnishingsand interior decoration and, in consequence, for composition ornament.)
Along with a proliferation of styles in the mid-to-late decadesof the century, there was a parallel growth in the number of moldableand castable materials that shared some features of the compositioncraft, such as carton pierre, gutta percha, fibrousplaster, shellac compositions and, eventually, celluloidand hard rubber. Composition continued to be the preferredmaterial for detailed decoration on wood where the size of theornament did not make its cost prohibitive. The publication ofpractical books by and for craftsmen, beginning in the 19th century,disseminated recipes and procedures to a broad audience and de-mystifiedthe craft. Period composition ornaments called "imitationwood carvings" were widely advertised in manufacturers' catalogs(see Fig. 14). Balls of prepared compo became available from someart supply shops in large cities for use by small volume craftsmen.
During the later years of the century, the Arts and Crafts Movement-aspreached by William Morris and his associates and followers-becameincreasingly important in design and philosophy. Morris stressedhonesty to the material in design, exalted spirituality of handwork and rejected manufacturing, mass production and the distinctionbetween "high" art and craft. These trends were to affectboth technology and design in the 20th century. Composition ornamentwould have been anathema to Morris and his elite clients; mostcomposition production during the last years of the century isbest described as Victorian Eclectic.
The 20th Century. The Arts and Crafts and relatedstyles, such as the more decorative Art Nouveau, were well rootedin America by the beginning of the century (see Fig. 15). Pitchmolds made from relief-carved patterns had become common in America.The carving tool marks could be accentuated in these patternsin keeping with current vogue. Open-grained
woods, such as mahogany, were often chosen so that the finishedcomposition ornaments would have a wood-like grain that showedthrough stains and varnishes (see Fig. 16). A uniquely 20th centuryapplication of composition ornament was in the lavishly decoratedmovie palaces of the Depression era (see Fig. 17).
As interest in architectural embellishments declined, particularlyas a result of the austere post-World War II styles, so did thecomposition trade. Many old firms went out of business and theirmolds were dispersed or destroyed. The few that remained concentratedon restoration projects or were sustained by diversification intoother materials. By the 1950s and 60s, composition as a materialand craft had been all but forgotten.
An upsurge in hand craft production that started in the late 60sand has continued to the present-as well as increasing interestin historic preservation-has led to the renewed study of old methodsand materials, including composition. The few manufacturers thatremain have seen a large increase in their business, and an increasingnumber of people recognize composition as a unique ornamentalmaterial and want to conserve, restore, or create it (see Fig.18).
To some degree, the longevity of historic composition ornamentis related to the ratio of ingredients in the original mix andto the skill of the craftsman in applying it. But it is far moredependent upon interior climatic conditions and the long-termeffects of heat and dampness on both the compo and the wood substrate.
Variables in mixing and application. Dried compois inherently hard and somewhat brittle; its increasing brittlenessover time is primarily due to the oxidation and hardening of thelinseed oil component. The drying oil, in turn, contributes toage cracking. Thus, during initial manufacture, if the oil contentwas low and the dry filler content (chalk) high, shrinking andcracking over time is less likely to occur. Originally, the compowas probably attached using small, headless brads (1/4")that penetrated the hardening compo as well as the wood substrate.They were used to keep the compo from shifting or warping afterit was set in place. If an insufficient number of brads was usedby the craftsman during the application process, the compo simplyfalls off as cracks develop.
Interior environmental conditions. Compo was conceivedas a durable substitute for hand-carved wood or marble and decorativeplaster; its potential for structural failure is generally dueto substrate failure rather than to the compo mix itself. Theoretically,composition will move with atmospheric changes due to the moisture-sensitiveglue component. Its breakdown typically occurs when the wood baseexpands and contracts at different rates than the compo duringextreme temperature and humidity fluctuations. Especially whenit is close to a source of heat, such as directly over a fireplace,compo develops fissures or shrinkage cracks. Contemporary heatingsystems in old buildings also contribute to the drying and crackingsyndrome.
Simple stabilization and repairs to existing ornamentation canmost likely proceed based solely on an analysis of existing conditions(see paragraphs on Surface Cracking andDelamination,below).
Historical research. For more complex work, a buildingowner, curator, or conservator should research the history ofthe building to find out when it was originally designed and constructed;who lived in it at various times; how the building was used; andwhich features were original and which were added later or removed(see Fig. 19). Some of this information may be found in the NationalRegister of Historic Places.
Questions about the building's interior spaces and their decorativedetailing also need to be asked, particularly when portions ofthe ornamentation will be replaced. Have the interior spaces evolvedwith successive occupancies or uses? In addition to compo, wereother decorative materials used and are there differences in patternsthat help date the work? For example, plaster and compo may havebeen used in the same room, but applied at different times. Receiptsfrom workmen's bills may often be used to establish the datesof decorative detailing.
The historical research dealing with the original constructionof the building and its use over time should, in turn, be linkedto the scope of work that will take place.
Stabilization, conservation, and repair are maximized within thetreatment, Preservation. Generally speaking, restoring decorativeornament to a specific earlier period is not recommended unlessits historical significance outweighs the potential loss of extantornament that characterizes other historical periods. But if asignificant interior is missing original features and physicaland documentary evidence are conclusive, replication may be appropriatein order to interpret a particular time.
Existing conditions analysis. After historical researchis conducted, but before starting work, an analysis of the surfaceand substrate should be undertaken. These are some of the issuesa conservator considers. First, if a surface is painted, the ornamentationmaterial needs to be identified. Is it wood, plaster, compositionornament, or some other type of applied ornamental material? Usually,some of the ornamentation is chipped or broken. Close examinationof the exposed material is the first step. If it is white throughthe entire thickness of the ornament, then it could be plasteror stucco; if it is a darker brown material, it is more likelyto be composition.
After having identified the presence of composition ornament,its overall condition can be evaluated. Layers of paint may obscurefine detailing as well as deterioration problems. Degrees of damageand deterioration should be recorded. These are typical questionsthat need to be answered. Is the surface merely "crazed",requiring no action or limited repair, or are the cracks severeenough to require replacement? Are pieces missing? Are the attachmentbrads rusted or missing? The condition of the substrate is alsoimportant. Is the wood surface intact, or is it in need of repair? After answering key questions,theconservator will make random tests to differentiate original compofrom later repairs, some of which may well have been done withplaster, rather than compo (see Fig. 20).
Deciding how to proceed depends upon the overall interpretivegoals of treatment. For example, is the interior being restoredto an earlier time? In this case, later repairs may be removedand the original appearance replicated. Or is the interior beingpreserved with limited replacement of lost or damaged historicmaterials? Not all conditions are foreseeable in conservationwork and contingencies must be incorporated into the treatmentplan to be considered realistic. As the project progresses, theconservator generally determines the work that needs to be done,and the order in which it should be undertaken.
The scope of work is generally based on several factors, includingthe historical significance of the building's interior, the degreeof damage or deterioration of the compo, and the overall interpretivegoals of project work. Several examples of repair and replacementfollow in order to suggest a typical scope of work within preservationand restoration projects. Treatments are listed in hierarchicalorder, from the least intervention to the greatest.
Paint removal. Interior ornament is usually paintedmany times over during its lifetime and, as a result, the sharpsurface detail of the original pattern is obscured. Before attemptingto remove paint, it is always advisable to obtain professionaladvice on the ornamental material to be cleaned as well as thenature of the coatings that are covering it. And whatever theproject work goal, at least one sample of intact, well-adheredpaint layers on a feature should be preserved for future historicalresearch.
Based on the purpose of treatment, these are some of the questionsa conservator routinely asks. How many layers of paint are there?Is it important to trace one layer to a particular occupancy ofthe building? If so, the stratigraphy (or layering scheme) willbe determined prior to paint removal. After the correct layeris identified, the color can be matched. Or, is the building beingrehabilitated? If this is the case, period-typical paint colorsmay be appropriate.
For purposes of this Brief and the guidance paragraphs that follow,it is assumed that all layers of paint are being removed in orderto reveal the fine detailing of the composition ornament (seeFig. 21).
The next step is to consider various methods of removing paintfrom the ornament without damaging it, or without being exposedto dangerous substances in the strippers or in the old paint itself!It should be noted from the standpoint of health and safety thatmost Federal and Empire period compo was meant to imitate marble;thus, the highly toxic white-lead paint was by far the most commonoriginal coating.
Caustic strippers based on lye should be avoided for tworeasons. First, they will damage and dissolve compo both becausethey "chew up" the protein structure of the glue and,second, because they are water-based and compo remains solublein water (see also Fig. 3). If a stripper will damage the proteinof your hands, it will do the same to compo!
A conservator will more often use organic solvents, suchas methylene chloride, in conjunction with small implements suchas a dental tool or toothbrush. (A small area is always testedfirst to establish the safety and effectiveness of any technique.Improper use of stripping tools can damage intricate surfacesbeyond repair.) A solvent is applied according to manufacturer'srecommendations, permitted to soak into and soften the paint,then re-applied as necessary, as the conservator gently removespaint from the intricate carved surfaces (see Fig. 22).
It should be emphasized that any amount of exposure to toxic chemicalswithout proper precautions can cause severe health problems. Ahooded, air-fed, personal unit is desirable when using methylenechloride-based strippers if fume hoods or paint spray booths thatexhaust effectively to the outside are not available. Organicvapor masks may not be as effective in protecting against methylenechloride exposure because the filters quickly become exhausted;however, a vapor mask with properly rated organic solvent cartridgescan provide an acceptable level of safety when cartridges areregularly changed (see Fig. 23).
Some conservators have had excellent results heat-stripping excesspaint layers using heat guns and dental tools. This ishighly skilled work and its success depends upon the compositionornament being much older than the paint layers that lie on top,but has the capability of working as well or better than chemicalmethods in the hands of an expert. Precautions must be taken againstlead fumes where removal of lead paint is involved.
Cleaning mixtures based on enzymes are also used by conservators.This is an effective method because enzyme mixtures can be formulatedfor very specific purposes (i.e., to dissolve only oil-based paintsfrom protein-glue based compo). They dissolve paint without affectingthe wood substrate. But, on the other hand, work can be very slowand the expense would only seem justified on small and rare orimportant museum objects. Enzymatic cleaners are dependent ona high level of skill, technical knowledge and professional training,but they are earning a solid place in the repertoire of professionalconservators.
Increased concern about the environment may well render the toxicmethylene chloride strippers obsolete in the near future. Manufacturershave already produced "safer" strippers based on dimethylesters, and further research will probably yield other alternativesto chlorinated solvents. Slower acting solvent-type strippersmay well be safer to the underlying composition ornament, butadditional research and use are needed before making definitivestatements.
In summary, most damage to compo occurs during the removal oflayers of paint; this is a critical process and should not beattempted without consulting a conservator and should not be undertakenby painting contractors unless they are highly skilled and havehad extensive experience in this very delicate procedure.
Proper disposal of residual chemicals and debris must be undertakento avoid contaminating the environment with solvents and lead,and such disposal is, in fact, now required by federal, state,and local ordinances. The company responsible for removing chemicalwaste should be licensed to dispose of it, otherwise the propertyowner may be held accountable if disposal laws are violated.
Refinishing compo ornament usually follows stripping. Accordingto historic evidence uncovered and depending on the existing anddesired appearance of the room, compo can be stained, painted,gilded, marbleized, or glazed. Paint types may include distemper,alkyd oil, or latex. A thin coating is recommended so the intricatesurface detail is not clogged.
Surface cracking. Surface cracking indicates ageand, thus, the history of the ornamentation itself (see Fig. 24).It does not necessarily mean that cracks have to be fixed. Butif cracking interferes with the overall design pattern, then theconservator may elect to fill the cracks with suitable fill material.For example, "light weight" spackles bulked with microballoonsare excellent because they are soft and compressible and willaccommodate changes in the size of cracks due to moisture fluctuation.After stabilization, the surface is finished to match the existingarea.
Delamination. Delamination or separation of thecompo from the wood substrate is the simplest repair problem toremedy. The conservator begins by testing cracked areas with slightfinger pressure to determine which parts of the design need consolidation.Compo sections that have separated from the substrate, but areotherwise intact, can be glued back in place using emulsion typeadhesives such as "white" glues or a clear, solvent-releaseadhesive (see Fig. 25). For vertical surfaces, the glue is paintedonto the back of the delaminated compo as well as the wood baseand, when slightly tacky, re-attached, and held with clamps untildry.
Professional conservators often formulate their own adhesivesbased on stable synthetic polymers (plastics) dissolved in solventthat will be more reversible, should the need arise, and alsooffer better long-term stability than many commercial adhesives.
Repairs to broken or damaged compo. When some originalcompo has been lost, additional work is required to make a repair.One particularly easy and inexpensive method of repairing brokenornamentation is to use non-hardening clay ("plastilina")or polymer-based modeling materials as an impression materialto make a mold. After a mold is made from existing ornament, missingor deteriorated portions of the historic design can be duplicatedwith a durable gypsum plaster (see Fig. 26). Especially in caseswhere economic considerations dictate procedure, use of this substitutematerial may be helpful because it is cheaper. Alternatively,an existing studio mold may sometimes be used to make small replacementpieces in a repair project (see Fig. 27).
In another scenario, a repetitive design on a mantelpiece maybe damaged or portions missing. Especially if the compo designis complex and several portions of ornament need to be replaced,rigid polymer molds with traditional compo are recommended forthe repair work. The mold is created using a section of the originalornament as a model. After replacement pieces are fabricated,they are attached using brads, or finish nails (see Fig. 28).The pointed end of the nail is clipped blunt with snips to avoidpossible splitting of the wood substrate. The nail is first hammeredinto the surface, then countersunk, and the resulting hole filledwith gesso putty or additional compo.
Finally, a ready-made replacement piece can be ordered from thecatalog of a compo manufacturer, but it is unlikely to be a perfectmatch to an extant historic decoration.
Replacement of missing compo ornamentation. Once-attractivecompo may become damaged to such a degree that the remaining fragmentsare removed by an owner and the entire surface painted over. Thus,if there is some existing composition ornament in a room, suchas an overdoor or chair railing, the conservator would most likelylook for evidence of other ornament that is now missing.
For example, a mantel may appear as a flat, unornamented surfaceto the untrained eye, but after many layers of paint are removedby the conservator, shadow images are revealed (see Fig. 29).These images or "ghost marks" are left by the hide gluecomponent of the original mix. Although the glue is water soluble,it will not be completely removed by an organic stripper suchas methylene chloride. (But if earlier inappropriate paint removalmethods were used, such as water-based strippers, caustic strippers,or mechanical sanding, ghost marks from the glue would be destroyed.)
When the paint stripper dries, a ghost mark left by compositionornament appears slightly darker than the surrounding area whereno compo had been attached. In addition, small, square-headed,¼" brads used to reinforce the original compo may beembedded in the wood.
In summary, detailed physical evidence, as well as written andpictorial documentation, can provide a valid framework for replacementat a particular site. With careful detective work, missing historicornamentation may be successfully identified and replaced withmatching ornament (see Fig. 30; see also Fig. 19).
Restoration of a "period" interior. Whenornamentation is extensively deteriorated and missing, ownersoften want to re-create the historic appearance through restoration.Physical evidence and other documentation may be used as a basisfor the restoration; it should be remembered, however, that asthe amount of surviving material diminishes, the greater the chancefor inaccuracy when attempting to depict the historic appearance.Choosing restoration as a treatment thus requires exacting documentationprior to work and meticulous attention to detail in the work itself.
Despite its popularity and widespread use as a decorative material,the history of composition ornament has yet to be thoroughly studied.Individual craftsmen have acquired fragmentary knowledge aboutsome designs and historic methods; historians and students ofinterior decorative design have accumulated knowledge about patterns,artisans, and methods of manufacture and distribution; and curatorsof historic collections that include compo are knowledgeable aboutthe objects under their care. The combined knowledge of theseindividuals, together with examples and images of compo ornamentfrom a variety of sources, needs to be synthesized to addressthe complex issues involving compo repair and preservation. Thefuture of the study of composition ornament, as well as many otherfacets of architectural, decorative, and fine art history, liesin this sort of cooperative effort.
Cover photograph: The process of making composition ornamenthas changed little over the years. In the J.P. Weaver Company,located in Glendale, California, freshly made compo is being kneadedprior to pressing it in a mold.
Adair, William. "An Investigation of Composition Ornamentation."The Interiors Handbook for Historic Buildings II. Washington, D.C.: HistoricPreservation Education Foundation, 1993, Chapter 4, pp. 1-7.
Adair, William. The Frame in America, 1700-1900: A Survey ofFabrication, Techniques and Styles. Washington, D.C.: The American Institute ofArchitects Foundation, 1983.
Budden, Sophie (ed.). Gilding and Surface Decoration. London:United Kingdom Institute for Conservation, 1991. See: Judith Wetherall, "History andTechniques of Composition," pp. 26-29; Jonathan Thornton, "Minding the Gap:Filling Lossesin Gilded and Decorated Surfaces," pp. 12-17.
Cotton, J. Randall. "Composition Ornament." Old-HouseJournal. Vol. XXI, No. 1 (January/February 1993), pp. 28-33.
Green, Malcolm. "Conservation and Restoration of Gilded Antiques."The Conservator, 3. United Kingdom for Conservation, 1979.
Hasluck, Paul N. (ed.). Cassell's Cyclopedia of Mechanics,8 vols. London: Cassell and Co., 1904, Chapter 4, p. 164.
_______________. Mounting and Framing Pictures. London:Cassell and Co., 1899.
Kunou, C.A. Manual of Gilding and Compo Work. Los Angeles, California:The Bruce Publishing Co., 1928. Request reprint information from the InternationalInstitute for Frame Study, 2126 "O" Street, NW, Washington,D.C. 20037.
Loeffler, R.F. Step by Step Compo and Mold Making. Oroville,California: Loeffler-Valac Industries, 1992.
Millar, William. Plastering Plain & Decorative: A PracticalTreatise on the Art & Craft ofPlastering and Modelling. London: B. T.Batsford. New York:John Lane, 1899.
Scott-Mitchell, Frederick. Practical Gilding. London: TheTrade Papers Publishing Company, 1905.
The Gilder's Manual. New York: Excelsior Publishing House,1876. Reprinted by the Society of Gilders, Washington, D.C., 1990.
The Secretary of the Interior Standards for the Treatment ofHistoric Properties. Washington, D.C.: U.S. Department of the Interior, National ParkService, Preservation Assistance Division. Washington, D.C., 1992.
Thornton, Jonathan. "Compo: The History and Technology of`Plastic' Compositions." Preprints of papers presented at the 13th annual meeting,Washington, D.C. American Institute for Conservation, 1985.
For information on conservators, contact the following organizations:
Association for Preservation Technology
904 Princess Anne St.
Fredericksburg, VA 22404
National Institute for the Conservationof Cultural Property
3299 K St., NW, Ste. 403
Washington, D.C. 20007
American Institute for the Conservation of Historic& Artistic Works
1400 16th St.
Washington, D.C. 20036
Kay Weeks, project director for this cooperatively producedBrief, is an art historian who serves as technical writer-editorin the Preservation Assistance Division. Jonathan Thorntonauthored the historical overview portion of the Brief and WilliamAdair, FAAR, the planning and treatment portion. The editorand authors wish to extend their gratitude to those people whoreviewed and commented on the Preservation Brief in draft formand those who provided illustrative materials. First, NationalPark Service staff reviewers included H. Ward Jandl, Blaine Cliver,Anne Grimmer, Chuck Fisher, Tim Buehner, Emogene Bevitt, Tom Jester,Michael Auer, and Paul Alley. Specialists in the field includedAndrew Ladygo, David Flaharty, Phil Gottfredson, Mark Reinberger,and Lenna Tyler Kast. Photographs were generously donated forthe Brief by Philip L. Molten, Elizabeth Brick, Robert J. Rucinski,Lenna Tyler Kast, Bryan Blundell, Thomas Brunk, Lonnie J. Hovey,AIA, Roland White, Irving Haynes & Associates, PhiladelphiaMuseum of Art, Cleveland Museum of Art, the Freer Gallery of Art,Winterthur Library, Decorators Supply Corporation, and Rapid Photography,Inc.
Washington, D.C. May, 1994
This publication has been prepared pursuant to the NationalHistoric Preservation Act of 1966, as amended, which directs the Secretaryof the Interior to develop and make available information concerning historicproperties. Technical Preservation Services (fTPS), Heritage PreservationServices Division, National Park Service prepares standards, guidelines,and other educational materials on responsible historic preservation treatmentsto a broad public.