Display techniques for museum objects are well covered in the literature (e.g., Witteborg), and specific information on instruments can be found in Eliason and Hellwig, and Barclay, Anatomy.

Figure 2. A mount for a flute showing support for the instrument along its length. Soft, resilient padding should be added at points where the instrument rests on the mount.

Figure 3. A typical mount for violin showing soft padding at points of contact, and fine monofilament security lines to prevent accidental slippage.

Chapter Head

Storage

• Can crowding of storage be minimized to limit handling of instruments during routine exami-nation?

• Are instruments easily accessible without needing to disturb others in front of them?

• Do the storage drawers run smoothly, or do sudden shocks when opening and closing cause specimens to roll or slide?

• Are handling techniques, or lack of control over handling, causing damage?

Storage conditions for museum objects are well covered in the literature (e.g., Barclay, Care; Johnson and Horgan).

Chapter Head

Handling

Polished metals are readily affected by incautious handling. The salts and oils from skin can cause corrosion of metals very easily, so always wear clean cotton gloves when handling metal instruments. Clean pairs of gloves should be placed in locations convenient to the storage areas as a reminder and inducement to handlers. However, gloves can snag on loose flakes, splinters, or corrosion, and they may cause highly polished and smooth objects to slip out of one's grasp.

Metal instruments should always be well supported during transport and display. Slides, mouthpieces and other potentially loose items should be secured. It is tempting to think that metal objects are very durable, but this is not always the case. For example, brasses that have been stressed during manufacture (i.e., sheet brass instruments) can be very weak, and can crack along the grain boundaries of the metal when put under further stress.

Organic materials deteriorate in a bewilderingly large number of ways, all of which potentially cause weakness. Leather bellows may be badly degraded and be easily damaged by one incautious inflation; apparently stable wooden instruments may be tunnelled by furniture beetles, and so on.

Below is an examination of various conditions, categorized by the substrate (i.e., the material of which the object is made), and the coating applied to its surface. Each of the conditions shown below can be assessed by eye without extensive training. Simple guidelines for cleaning, handling, storage, and display are indicated in the key.

Up

Key to Handling Guidelines

2. Fibrous surfaces are very attractive to dust and difficult to clean. Keep such items well covered, or in boxes or drawers.

3. Flakes can catch on gloves or clothing. Handle as little as possible. Place the instrument on a support for carrying and examination. Some form of consolidation may be necessary if there is evidence of deterioration.

4. Porous objects may be perfectly stable, but their surface can absorb dirt and moisture. Never wet clean a porous surface. Keep the items well covered, or in boxes or drawers.

5. Powdery surfaces may be very fragile. Handle as little as possible. Place items on a support for carrying and examination. Some form of consolidation may be necessary if there is evidence of deterioration.

6. Rough surfaces can catch on gloves or clothing. Handle as little as possible. The rougher the surface, the more likely dust and dirt will stick. Keep covered.

7. Smooth objects should be handled with soft gloves, although glass and ceramics can slip if not handled carefully. Cleaning by gentle dusting is usually very simple.

Chapter Head

Travel

Chapter Head

Strategies to Counter Biological Attack

Insects can be affected by inert atmospheres in two ways: they may suffocate and they may fall into narcosis. When nitrogen is used, the insects die of suffocation because of insufficient oxygen. There must be very little oxygen in the nitrogen atmosphere for the process to be successful (0.1% or less), which makes this procedure technically demanding. On the other hand, placing insects in a carbon dioxide atmosphere has the effect of stimulating the insects' respiratory receptors causing the fine balance between carbon dioxide and oxygen to become upset, resulting in narcosis. The effect takes place in a relatively low carbon dioxide concentration, so there is no need to exclude oxygen entirely from the atmosphere. Thus, the process is easier to apply and less time-consuming and costly. Small instruments are placed in a chamber filled with carbon dioxide from a tank of the compressed gas. Larger objects may be enclosed in a sealable polyvinyl tent or chamber. A level of approximately 60% CO2 must be maintained during the treatment. Time of exposure depends upon the species of the insect, and its development phase. The Canadian Conservation Institute is planning a publication on this topic.

Preventive cleaning and regular inspection forestall insect attack. Access to many parts of musical instruments for cleaning is extremely awkward and sometimes demands disassembly. This is especially so with keyboard instruments. In the case of keyboard instruments maintained in playing condition, cleaning should only take place infrequently, when another problem warrants the disassembly, or when there is a reasonable suspicion of infestation. The cleaning process itself should be given plenty of time, so that adequate care can be taken. If infestation is found, it may be necessary to seek specialist advice (Pinniger; and Dawson and Strang, Insect Problems).

Fungal attack occurs in areas of high relative humidity, above approximately 65%. The popular belief that fungal growth requires elevated temperatures is not true. Spores may lie dormant on organic materials for long periods of time and only become active when conditions are right. The best preventive measure is to lower local relative humidity, especially during periods of dampness, check for damp spots, move instruments into areas of higher air flow, and establish programmes of regular inspection (CCI Notes 8/1; Dawson and Strang, Fungal Problems).

Larger musical instruments are an ideal habitat for vertebrate pests, especially when they are in storage or unattended. Mice, bats and other such pests are better dealt with by trapping, blocking the means of access, and establishing programmes of inspection and regular cleaning (Dawson and Strang, Vertebrate Pests).

Chapter Head

Return to Contents

This chapter introduces some of the chief materials used in musical instruments. It describes briefly methods of fabrication, working characteristics, and the causes of deterioration. The conservation treatment of the materials described here is covered in Chapter 5. Among the many publications on the materials of artifacts, An Introduction to Materials, in the Science for Conservators series, and The Organic Chemistry of Museum Objects, by Mills and White, are especially recommended.

Chapter Head

Metals

Metal Production

Exceptionally stable or well-protected metals are found pure in nature. The metal is referred to as "native" in this case. Silver and copper are sometimes found in the pure state, while gold always occurs in the metallic form. The remaining metals are found as ores. They are "oxidized", but not necessarily with oxygen. The term "oxidized" refers to the chemical process. Metals may be combined with many different elements and compounds. The oxides are most common for metals like iron and aluminum, while copper may be combined with sulphates, carbonates and so on.

Once mined, these ores must be refined, and because the metals are at their most stable, lowest energy state, an input of energy is required to refine them. This is done by heating the ore under certain conditions until the fluid metal becomes separated from the slag. This is the process known as smelting.

Many metals in artifacts are composed of alloys, which are mixtures of two or more metals. This mixing has the function of enhancing the usefulness of the material by combining the properties of individual metals. For example, zinc and copper are combined to produce brass. Iron is mixed with a non-metal, carbon, to produce steel.

Metals have three specific properties as elements:

• malleability

• ductility

• conductivity

Malleability enables metals to be worked into shape by beating. Ductility enables metals to be drawn into wires or rolled into sheets. The facility to conduct electricity is not greatly important to conventional historic instruments, but it has significant effects on the processes of corrosion. One might also say that metals are machinable, but when the process of cutting at the tip of a tool is examined it is evident that a combination of malleability and ductility allows the metal to "flow away" from the cutting tool. Malleability and ductility may be greatly modified by alloying. The Nature of Metals by Rogers provides a solid introduction to this topic, while Barclay provides a synopsis in The Art of the Trumpet-maker.

Up

Fabrication

• Forming
  • Casting, forging and rolling where ingots of metal are given their initial shape.

• Shaping
  • Milling, turning, spinning, grinding, stamping, cutting, and drilling, where the final dimensions of the piece are established.

• Assembling
  • Welding, soldering, brazing, riveting, bolting, crimping, and shrinking, where pre-made parts are put together. It is also worth noting that adhesives have been used occasionally to join metal parts.

• Finishing
  • Plating, burnishing, polishing, etching, painting, lacquering, engraving, chasing, embossing, enamelling, and patinating, where the piece is finished for decorative purposes or for further protection.

Instruments or components routinely pass through all of these stages and as the number of processes applied multiplies, so does the difficulty of deciding in retrospect exactly what was done to them originally.

Up

Corrosion

Figure 4. Cross section of a metal surface, showing the negative charge and the reaction with positively charged ions in the atmosphere.

Oxidation can occur in dry conditions at the metal surface. A fresh surface is especially reactive and forms bonds with available electron recipients in the surrounding atmosphere. If a strong, even film is formed, it will protect the metal against further corrosion because it forms a barrier against other electron recipients. More active forms of corrosion need water, in addition to reactive agents in the air. Aqueous corrosion is an electrochemical process in which water, particularly "impure" water containing dissolved chemicals, acts as an electrolyte (conductor).

Some metals are more prone to corrosion than others. Metals are arranged in order of reactivity or potential for corrosion in what is known as a galvanic series. The metals at the lower end are regarded as base, while those at the upper end are called noble. In practical terms, if two metals are in contact (e.g., as an alloy, or a plating, or as soldered or joined components), the one lower down in this series (the more base metal) will corrode before the metal higher in the series (the nobler metal). For example, if an iron object is suspended by a copper wire in an environment containing sufficient humidity, the iron will corrode at the point of contact with the copper. The further apart the metals are on this scale, the greater will be the reaction between them. Such a reaction may also take place within alloys, which are a mixture of two metals. Conditions such as access of oxygen, concentration of solution, irregularities in metal, etc. cause accelerated corrosion. This "bimetallic" corrosion can cause confusion in identifying metals. In an alloy, the baser metal corrodes first, causing a deposit of its characteristic corrosion products on the surface of the artifact. Silver alloyed with copper may develop a green crust; brass may appear copper-coloured due to "dezincification."

The Corrosion Environment

For corrosion to take place the main ingredients are water, oxygen and salts, although all three are not required for a reaction to take place. Of the salts, the most destructive are the chlorides which, unfortunately, are very prevalent, and highly soluble. It is extremely difficult, if not impossible, to render metal stable in presence of chlorides. Both copper and iron combine with chloride ions to form corrosion products, which are unstable above approximately 35% relative humidity.

The rate at which corrosion proceeds is temperature dependent; it will proceed faster at elevated temperatures, more slowly at reduced temperatures. For example, silver tarnishing occurs twice as fast at 29^oC than it does at 18^oC. The destructiveness of corrosion products depends upon whether the corrosion products are stable, and whether there is a change in volume during the corrosion reaction.

Up

Passivation

Up

Corrosion of Specific Metals

Copper causes fewer problems in musical instruments than iron, but is still found in a wide range of corrosion states. It combines with chlorides, sulphur compounds, carbonates, nitrates, etc. to produce a wide range of colours and textures, some of which are sought as patinas. Oxides are red or black; carbonates are green or blue; chlorides are either waxy grey-white or light green and powdery; and sulphates are blue or green (CCI Notes 9/1 and 9/3). Generally, copper corrosion products form even, coherent layers that often preserve the details of the original metal surface. However, as with iron, chloride ions are the most reactive with copper, forming a white, waxy layer that is extremely unstable. When combined with oxygen and water this layer rapidly converts to a light green powder with loss of all detail and structure. Sometimes known as "bronze disease," the corrosion continues under favourable conditions until the metal is consumed. Copper alloys, such as brass and bronze, are susceptible to corrosion of the least noble of their elements. Brasses are particularly prone to stress cracking or "season cracking", where weakening and embrittlement are caused by inter-granular corrosion. This is often stimulated by exposing the brass to atmospheric pollutants, particularly ammonia. Brass is made of copper and zinc. Sometimes brass can lose its zinc preferentially to copper at the surface. This is known as dezincification and makes the brass appear much more coppery. It is often a result of extreme chemical cleaning with acids. This surface enrichment, as it is called, can confuse identification of the metal.

Silver is relatively noble and less susceptible in general, but it does combine with chlorides and sulphides. Silver sulphides cause the classic blue-black iridescent tarnish, while silver chlorides are white initially, but are light-sensitive, and on exposure turn deep purple-grey. Chloride films are unlikely to be found in indoor conditions, and are usually only encountered on archaeological material. Sulphide tarnish, on the other hand, is a very common museum and household problem. Because people have a low tolerance for tarnish on silver, instruments are vulnerable to physical damage caused by abrasion and loss of surface when polished (CCI Notes 9/7). Silver is often used on instruments as a thin plated layer over a baser metal. If the baser metal is exposed by scratches or polishing, bimetallic corrosion may occur.

Lead and its alloys are used in instruments for their weight and softness. When lead corrodes slowly in pure air, it forms lead oxide, a dull grey passivating layer. In the presence of impurities, it can form other oxides (red, yellow, or brown), sulphates (dark grey or black) or carbonates (grey-white or white). Most of these products are relatively stable, but lead has a strong reaction with weak organic acids (acetic, formic, etc.) and forms a loosely adherent, whitish powder. This is basic lead carbonate, an active form of corrosion. Some building, display, and storage materials can cause this because they are sources of weak organic acids. Lead is often alloyed with other metals, particularly tin, for use in organ pipes or as a solder for components made of other metals. Lead weights inserted into keys or jacks as balances are prone to corrosion due to their close contact with wood (CCI Notes 9/1).

Chapter Head

Wood

Up

The Structure of Wood

When describing wood structure, three planes ­ cross-sectional, tangential, and radial ­ are distinguished. Unless the orientation of the surfaces of the wood lies accurately in one of these directions, determining its species can be impossible. This is especially true when the wood is viewed under a magnifying lens or with a microscope.

Figure 5. Transverse, tangential and radial cuts of wood.

The structure of softwoods is simpler than that of hardwoods because they appeared earlier in the evolution of plants. The predominant features of softwoods, are visible to the naked eye, and appear more clearly through a magnifying lens.

Figure 6. Microscopic features of a typical softwood. Note the difference between late and early growth, and the resin canals. (Adapted from K. Mägdefrau.Botanik: Eine Einführung in die Pflanzenkunde. Heidelberg: Carl Winter Verlag, 1951.)

Under a 10-power hand lens, the resin canals characteristic of spruce (Picea), larch (Larix) or Douglas-fir (Pseudotsuga) will show clearly in a good cross-sectional cut. The presence of resin canals distinguishes the above species from such woods as fir (Abies), true cedar (Cedrus), cypress (Cupressus) or yew (Taxus). Botanical names are very helpful in clearly describing a wood species. The common names of many woods are quite misleading. For example: the northern white cedar and western red cedar are not cedars, but members of the genus Thuja. The eastern red cedar belongs to the genus Juniperus, which is a member of the family Cupressaceae, and so on.

The structure of hardwoods is differentiated by a larger number of specialized cells. The presence of vessels characteristically distinguishes softwoods from hardwoods. Vessels are clearly visible in such woods as oak, balsa, plane and chestnut. In other species, they may be so small that they are hard to detect even with the aid of a lens, as is the case with boxwood. When the vessels are distributed evenly throughout the cross section, the wood is known as a diffuse-porous hardwood; when the vessels appear grouped in annual rings, the wood is known as ring-porous hardwood.

Figure 7. Microscopic features of a typical hardwood. Note the large diameter vessels clearly visible in the cross section. (Adapted from K. Mägdefrau. Botanik: Eine Einführung in die Pflanzenkunde. Heidelberg: Carl Winter Verlag, 1951.)

Annual rings appear in both softwoods and hardwoods growing in climatic zones that experience seasonal changes of both temperature and rainfall. In spring, when there is an abundance of water, large cells with wide cell lumina (openings) are produced, then as the ground dries later in the year the cells become more substantial with thicker walls and smaller lumina. These annual rings determine the appearance of many wood species and allow a determination of the tree's age. However, the brown or black stripes observed in rosewood (Palisander or Dalbergia) are deposits of coloured material in certain stages of the tree's growth, independent of regularly occurring climatic changes.

Another important feature in both softwoods and hardwoods is the formation of rays ­ cells running radially between the centre of the tree and the bark ­ that contribute much to the visual appearance of wood. Typical examples of woods with pronounced and easily visible rays are beech, oak, and plane. Rays also play an important role in the swelling and shrinking of wood. To identify the species of wood, these and many more details are used in combination. In some cases, only a microscopic analysis can reliably identify the species in question. Hoadley's two publications on identifying wood are recommended.

The various anatomical features of wood, together with physical properties like density and chemical make-up, and the inclusion of extractives or solids, all determine the texture, colour and odour of the wood, as well as its hardness, resistance to insects, fungus, and physical decay, gluing properties, and general workability.

Up

Identifying Wood Species

While a non-specialist can examine some species of wood successfully with a hand lens, microscopic determination is much more challenging, and needs special training, continuing practice, and costly equipment. The quantity of wood needed for microscopical examination can be quite small ­ samples 1 sq. mm and 0.02mm thick are more than sufficient ­ but such samples must always be taken by the person who is doing the microscopy. An immediate check of the sample is necessary to ensure that it is perfectly oriented, as well as thin enough to show all the required details.

Up

Wood Species Used in Musical Instruments

Up

Wood and Moisture

The changing amounts of water vapour adsorbed on the surface of the cell walls determine the dimensional changes of wood. When wood is at equilibrium with the air around it, moisture may be given off to the surrounding atmosphere, or it may be absorbed from the atmosphere into the wood. Thus, conditions in the environment determine the moisture content of wood, and thereby its swelling and shrinking. Moisture content in relation to relative humidity in the air is shown in the graph.

Figure 8. The relationship between relative humidity, temperature and moisture content in wood.

This graph provides average values and needs correction for individual wood species, especially in the higher range of relative humidity. For the middle range of temperature and humidity, however, this graph is quite adequate. Wood shows a different reaction to moisture changes along each of its three axes ­ longitudinal, tangential, and radial. The ratio of reactivity between the three directions is:

Longitudinal = 1

Radial = 10

Tangential = 20

Thus, for every unit of change in the longitudinal direction, there is 10 times as much change radially, and 20 times as much tangentially. In practice, the longitudinal changes are small enough that they may be ignored. Thus, with humidity fluctuations in the surrounding air, there is little or no longitudinal change, and the tangential response is double that of the radial. The actual amount of shrinkage or swelling depends on the density, the degree of coloured heartwood formation, and the quantity and kinds of extractives, to name only the most important factors.

The following table provides average percentages for dimensional change with a 1% change in moisture content of various woods:

To calculate an example: a cube of oak of 100mm on a side is in equilibrium with 20^oC and 55% relative humidity. The RH is then lowered to 40%, which leads to a decrease from 10% to 7.5% moisture content once the new equilibrium has been reached. This decrease will result in negligible longitudinal shrinkage, and shrinkages of 0.5% radially and 0.8% tangentially. This corresponds to a decrease of 0.5mm radially and 0.8mm tangentially. Such small figures may not seem important until one considers the back of a viol or the soundboard of a harpsichord. When the above calculation is performed using the figures for maple and spruce respectively, including the width of the wood in question, the results indicate the large extent of movement.

Such dimensional changes take place only if the piece of wood is unhampered while the alterations in its environment are taking place. For restrained pieces of wood, there are two scenarios depending upon loss or gain of moisture:

• If wood is losing moisture, the forces of shrinkage may be stronger than the internal cohesion of the wood. This will result in open cracks. However, not all such forces lead to fracture because wood possesses an elastic quality that minimizes the formation of cracks. (This effect, due to plasticity and elasticity, will be discussed in the next section.) As an example, a violin or viol top that is glued all around its edges may undergo moisture loss without the formation of cracks.

• If the dimensional changes are in the opposite direction, where the wood is swelling due to a gain in moisture, this will lead to a structural change at a microscopic level. The cells become compressed. Then, when the wood is subsequently dried, the compression will become apparent in permanent shrinkage and the formation of cracks. This effect is called compression shrinkage, and the damage can be ob-served in many wooden constructions where flat pieces of wood are restrained in their movement by cross-glued bars. The resonating surfaces of guitars, viols, or keyboard instruments often show such signs.

  Theoretical values for dimensional changes often differ from measured values when surface coatings, such as paints and varnishes, have been used because they prevent or inhibit free exchange of moisture with the air.

Up

Plastic and Elastic Behaviour

Wood is also elastic. This term describes the behaviour wood exhibits after it has been under stress. A string instrument bow is tightened for playing yet returns to its original shape after the hairs are detensioned (for example, see Figure 1).

Figure 1. A typical stress/strain curve.

Woods used for bows are chosen specifically for their elastic properties. Plasticity and elasticity of a particular wood depend upon its moisture content. Maximum plasticity is reached once all fibres are saturated by water (fibre saturation point). Minimum plasticity occurs when wood has been dried in an oven and thus deprived of all its moisture. In musical instruments plasticity and elasticity are very important. Plasticity may prevent cracking, while elasticity maintains the intended shape and particularly the vibrational properties. On the other hand, plasticity is responsible for any deformation that will take place even under the most strictly controlled conditions.

Environmental Requirements

When wood is dried after it has been felled in the forest, it loses moisture until it is in equilibrium with its surroundings. The wood is then described as dried, or seasoned. Because this depends on the ambient conditions, dryness is a relative term. An object made of wood cut, dried, worked, and stored in the Sahara region will have a very different moisture content than an object from the damp and foggy Po region of Northern Italy. To prevent further loss or gain of moisture, and subsequent dimensional changes once the object is transferred to a museum, the object's original climatic conditions should ideally be maintained. However, in practice this is hardly feasible, and in many cases not even desirable, because an instrument will most probably already have adapted to some other humidity level. Also, the presence of other materials on wooden musical instruments make absolute specifications difficult to establish. For storage or display purposes, a compromise climate has been recommended (Karp, "Storage Climates").

Up

The Aging Process of Wood

Visually, on exposure to light, coloured wood slowly fades, while woods originally light in colour tend to darken. Thus, on an instrument or any other wooden object exposed to light, yet protected from weathering, all the colour shades of the woods will eventually meet somewhere in the middle. While the overall effect can be very attractive in many stringed instruments or woodwinds, it is very disturbing in the case of marquetry or inlays made of differently coloured or dyed woods. These slow colour alterations indicate chemical changes on the surface of the wood, but they have no effect on an instrument's tonal qualities.

Few investigations into the ageing process on the physical and mechanical properties of dry wood have been made. There are many examples of dry wood that has been protected from biological attack by fungi and insects for thousands of years. While some studies seem to indicate that the density of wood increases with time, other research suggests the opposite. Certainly, the rate of water absorption is apparently un-changed over time. It has recently been shown that under continuous forced vibrations, the stiffness of wood is increased and the damping coefficient decreased, but such observations are difficult to interpret because the physical and mechanical data of wood species generally show such wide variation. In addition, one cannot discount the subtle and indefinable interactions that take place between the musical instrument, the musician, and the auditor. To simplify, it is a safe assumption that the properties of dry wood alter very little, if at all, with time.

Up

Wood and Musical Properties

It is, however, a mere myth that instruments "lose their voices" unless they are played regularly. Instruments may indeed lose some of their easy response if not played for a longer period of time, but the tonal qualities are easily regained once the instrument is adjusted and settled to playing condition. Thus, the so-called "maintenance" that the splendid violins on display in the City Hall of Cremona receive by being played every day, contributes to the risk of damage, rather than to the upkeep of their musical value.

Up

Choosing Wood for Repairs

The shortcomings of repairs to musical instruments usually relate more to the quality than to the kind of wood used. Some conservators prefer to use well-seasoned wood that has acquired a brownish colour, matching the colour of the original. Orientation of the piece to be inserted is important. Earlywood and latewood in the original and repair must lie in the same direction and the angle between the longitudinal fibre axis and the wood surface must also be matched.

Chapter Head

Skin Products

Skin Products

Skin is composed of four layers; the epidermis on the outside, the dermis just below this, the corium which comprises the body of the skin, and the hypodermis, which lies on the inside. The inner and outer sides of skin products often show two characteristic textures. During the preparation process, the epidermis and hypodermis are usually removed, leaving the smooth-sided dermis, and the fibrous-sided corium. Thicker skins are often split during processing, and various textures can also be applied, so the inside and outside of the skin may not be obvious.

Figure 9. Cross section of skin, showing the epidermis, the grain layer and the corium.

A wide range of techniques have been applied to processing skin products. Among the finished materials are rawhide, parchment, semi-tanned skin, smoke-tanned skin, chamois (fat liquored), alum tawed skin, and vegetable and mineral tanned leathers. In addition, many special treatments have been applied to the finished product including boiling (cuir bouilli), patent leather, texturing, etc. The chief products used in musical instruments are rawhide and parchment (untanned skins) and leather which has undergone a chemical tanning process.

Rawhide and Parchment

Once skin has passed through a pre-treatment process, but before being turned into leather, it can be used for making artifacts. In this state it is known as rawhide. Rawhide can be used in a number of places where its shrinkage as it dries can be used. For example, wooden components may be covered with rawhide in its wet state so that when the hide dries it exerts powerful contracting forces, pulling the wood joints tightly together. Similarly, drum heads can be made of wet rawhide, which tightens on drying to produce a taut membrane.

Rawhide is very susceptible to moisture, and if wetted sufficiently it will putrefy. Because no tanning process is used, the collagen molecules are still reactive to water. In its normally dry state it is extremely tough and intractable, and may fracture if bent or twisted. Wetting and subsequent drying will cause very high shrinkage.

Parchment is prepared from pre-treated skin that is still in the wet, swollen stage. The critical stage comes when the wet skin is stretched on a frame while allowed to dry. Unlike the process of making leather, where the skin is not under tension, the individual fibres are stretched during drying. This causes them to pack more tightly to each other and allows them to be held in place when the matrix material dries. Thus, having dried under tension, the skin is hard, flat and relatively thin. The skins of small animals are generally used for making parchment, particularly calf, goat and sheep. Sheepskin parchment is the most common. In order to whiten the skin and make the surface smoother, various fillers like chalk, talc, and whiting are rubbed in after the skin is degreased. The surface is sometimes treated with an abrasive, such as pumice or brick dust, to even out the texture.

When stored below 50% relative humidity, parchment is very durable. However, higher levels of RH may cause it to shrink and cockle very badly as the stretched collagen fibres attempt to regain their original shape. Parchment can also embrittle with age causing stretched areas to tear.

Fish skin is used in some wind instrument paddings. It is high in keratin and has excellent flexibility and good resistance to moisture.

Up

Leather

Most of the world's supply of leathers is now mineral tanned, although the process is only about a century old. Thus, mineral-tanned leathers may be found in more recent instruments, but the leather on the majority of historic instruments will be vegetable tanned. Chromium salts are predominantly used in mineral tanning, and the process is thus known as "chrome tanning." Other salts, particularly zirconium, have recently come into use.

Vegetable tanned leathers are generally quite stable, but in musical instruments they are most often used in places where flexibility is necessary, such as the hinging of piano hammers. Organ bellows combine flexibility with air-tightness. In such high use locations, old leathers can become mechanically deteriorated through constant flexing, and the materials used by the tanner to soften the leather may also dry out and become ineffective. Some vegetable-tanned leathers suffer from red rot, which is a deterioration brought on by exposure to sulphurous pollutants in the atmosphere. In extreme cases of red rot, the leather becomes light brown and very powdery, and disintegrates.

Chapter Head

Bone and Ivory

Figure 10. Microscopic features of bone, showing the Haversian system much magnified for clarity.

When bone is cut and shaped it is the Haversian canals that give it its characteristic appearance. On some very dense structural bones, this pattern may not be noticeable to the naked eye due to its very fine structure, so bone can be mistaken for ivory. Even though it contains a relatively high proportion of organic material, bone is not very sensitive to moisture.

Ivory is the teeth of mammals and is formed in the soft tissues of the jaw. A crown of enamel is laid down by specialized cells, and it takes on a honeycomb appearance. Apatite is laid down by the cells, as in bone, except that here the cells are not incorporated into the crystal structure. As the crystals form, the organic portion is squeezed out to the sides. This accounts for the comparative density and hardness of tooth material compared with bone. In a fully-formed tooth the organic portion is around 1%. In cross section, a differentiation is visible between the enamel on the upper surface and a second material, dentine, which forms the body of the tooth from the crown to the root. The root is held into the bone of the jaw with a third substance called cement.

Figure 11. Cross section of ivory tusk, showing the enamel (the outer layer) the dentine (the body of the tusk) and the cement that holds it in place in the jaw.

Elephant tusks have been the major source of ivory in the making of musical instruments, although narwhal tusks and teeth from other animals have also been used. Ivory is used for wear-resistant surfaces such as key facings, for decorative inlays, and for lathe-turned parts of wind instruments. Occasionally whole instruments have been made from it, especially woodwinds. Elephant ivory has a characteristic growth pattern that makes cut pieces of it easy to recognize in cross section (CCI Notes 6/1). Like bone, it is comparatively resistant to moisture, although because it is more brittle, cracking is more commonly observed. Both ivory and bone yellow with age, especially in areas that are not exposed to light.

Chapter Head

Keratinous Materials

Horn

Figure 12. Cross section of a horn, showing the horn as an outgrowth of the epidermis, and the interior bone

Because horn is a relatively plastic substance, from very early times it has been worked into various shapes. It is heated by boiling in water until it becomes flexible enough to form into quite complex shapes. In the 19th century, techniques were developed for grinding horn to a powder in strong alkalis, then re-forming it with heat and pressure. The resulting artifacts in no way resemble the original form of the horn.

Horn is used occasionally in the bodies of musical instruments, such as the traditional hornpipes, and is also used as an inlay for decorative purposes. It is very resistant to moisture, but can delaminate internally and crack with age.

Up

Quill

Figure 13. Detail of a feather, showing the barbed construction, and the parts used for making plectra.

Like horn, quill is a flexible, water-resistant and fairly durable material, but as it dries and loses natural oils it becomes brittle.

Up

Baleen

Chapter Head

Glass

Chapter Head

Synthetic Organic Materials

Cellulose nitrate (or nitrocellulose) is found under a variety of names including celluloid, french ivory, and collodion. It can be made to look like ivory, tortoiseshell, horn, nacre, and many other organic animal materials. In musical instruments it is usually found in synthetic key facings, which have been grained and figured to look like ivory. It contains a volatile plasticizer, camphor, which can diffuse through the plastic and vaporize from its surface. This leaves the material brittle and subject to shrinking and cracking. In extreme cases, the plastic crumbles into small cubical fragments. Nitrocellulose is notoriously unstable, especially when prepared for such applications as photographic film, and in a highly nitrated form it is used as a high explosive. While the material used for keyboard facings and similar parts is relatively stable and non-reactive, it remains highly flammable and can degrade badly under some conditions (CCI Notes 15/3).

Ebonite has been used since at least the middle of the 19th century for making woodwinds. It is a highly vulcanized natural rubber created by adding a large proportion of sulphur under heat and pressure. It is subject to deterioration, producing acidic exudates, which can attack the fittings of the instrument or other associated materials. Moisture stimulates this reaction (Bacon).

Chapter Head

Coatings

A coating can be applied to a surface in the form of a fluid, it may be a solid which is adhered to the surface, or it may be a metal deposited electrically from a solution of salts. The fluids may be single substances, like oils or waxes, or they may be solutions, emulsions or suspensions. Solids, such as wood, metal, paper, or leather, may be applied to a surface as veneers or inlays. Surfaces may also be modified by applying chemicals. Nine basic categories of surface coating are defined:

• pigment

• paint

• lacquer or varnish

• oil or wax

• veneer

• paper, leather or textile

• enamel or glaze

• plating

• patination

Although the following notes refer mostly to substrates made of wood and metal, many other materials might have similar coatings. For example, leather can be painted or lacquered, ceramics can be glazed, and many other materials may have their surfaces modified in various ways.

Up

Pigment

Usually, porous surfaces such as wood and other fibrous vegetable materials are coated in this way because the pigment adheres easily and is not rubbed off quite so readily with use. Movement of the underlying material (usually wood) has little effect on loosely bound pigments, because they do not form an intact film, but are scattered over the surface.

Up

Paint

Painted surfaces are easier to maintain than those covered in dry pigment, provided no deterioration has taken place. However, complete films on objects that expand or contract can crack and loosen. This problem is mostly seen on wooden objects and paintings on canvas, but paints on metals which have been heated and cooled may also show the same symptoms. Adhesion between the paint and the surface may also fail with time, and cause flaking and loss. In addition, the binders in paints may undergo deterioration, primarily due to the effects of light and oxygen, which can cause breakdown of the film, powdering, and changes in colour.

Up

Lacquer or Varnish

Lacquers and varnishes are sensitive to light. On exposure, they can become degraded, exhibiting yellowing, cracking, and loss of adhesion. As with paints, complete films applied on objects that expand or contract can become cracked and loosened. This problem is mostly seen on wooden surfaces, but transparent coatings on metals that have been heated and cooled may also show the same symptoms. Breakdown of lacquers on metals can cause spotty local corrosion.

Up

Oil or Wax

Surfaces that have been treated with materials that only dry slowly, or not at all, tend to collect accretions with use, and to collect dust when not handled regularly. In general, such surfaces are durable and stable, except where the coating has been applied very thickly, or where the surface is smooth and lacks adhesion. In the latter cases, the oil or wax may flake away leaving lighter-coloured areas. Such finishes are usually encountered on ethnographic instruments, but bare wood in traditional European workshops was sometimes treated with linseed oil or wax as a protective measure.

Up

Veneer and Inlay

Where two differing materials are in intimate contact, physical and chemical reactions between them are possible. Veneers on musical instruments are prone to splitting and loosening due to movement of the underlying wood, either by changes in relative humidity or by distortions due to tension. Metals used in contact with wood may corrode due to the hygroscopic property of the wood and the presence of corrosive extractives.

Paper, Leather or Textile

Paper, leather and textile are grouped here because their use on the surfaces of instruments tends to be similar. Paper on instruments takes two forms: decorative linings covering large surfaces, and paper labels applied in discrete spots. Paper is usually applied to the surface with pastes made from starches, although hide glues or gums may also be used. Leather is used as a covering and reinforcement, particularly on wooden wind instruments that have been pieced together. Cornetti, serpents and oboes da caccia often have a thin leather covering attached with hide glue. Textiles are sometimes used as reinforcements for wooden constructions; for example the linings of string instrument resonators.

Paper, leather and textiles are all very fragile and subject to wear, fading and embrittlement. On wooden objects, fluctuation of relative humidity can cause the covering to tear as the wood expands and contracts. These materials, and their adhesives, are also very susceptible to moisture. Exposure to light can cause embrittlement of paper and textiles, and the fading of pigments used on them.

Up

Enamel or Glaze

Enameled or glazed surfaces can suffer from crazing (a very fine cracking) due to differential movement of the substrate. Often crazing is incipient, meaning that it is a feature of the surface and unlikely to worsen unless subjected to poor conditions. Crazing is often valued as a decorative feature of surfaces. Enameled metal can show flaking if it has been stressed or distorted, or subjected to fluctuating temperatures.

Up

Plating

Adhesion between layers of metals can fail, especially where corrosion has occurred, or where the surfaces were not cleaned sufficiently before plating. Galvanic corrosion can take place between dissimilar metals in the presence of salts and water, especially where gaps in the plating exist. Thin platings are particularly sensitive to polishing and may be worn through very easily in high spots.

In surfaces where gilding has been applied to a non-metal, particularly wood or gesso, there is a risk of great fragility. Movement of the substrate can cause cracking and loosening. Gold, tin and silver leaf applied to such surfaces are also very thin, and susceptible to damage by abrasion.

Up

Patina

Because patinations on metal objects are chemical features of the surface atoms, they are usually very thin, and they are therefore susceptible to abrasion and chemical attack. Naturally developed patinas are often aesthetically valued and may also provide information on the use of the instrument.

Chapter Head

Composite Objects

The following notes refer primarily to the interactions between wood and metal, but many of the principles are common to composites of metals with other organic materials, and other combinations of materials.

Up

Metals and Wood

In common with other reactive metals, iron is most stable in dry conditions. Water is the chief initiator of corrosion reactions, so the drier the metal is, the slower will be any chemical reactions. However, cellulose, of which wood is made, readily absorbs water. If the cellulose has excess water, corrosion reactions may occur with any metal in contact with it. On the other hand, the presence of metal ions in aqueous solution (from corrosion reactions) causes deterioration of both the cellulose and lignin components of wood.

Copper and its alloys cause similar deterioration. Copper salts are more damaging to wood than those of iron. In addition, copper can react with organic acids in wood and any oils used in its treatment, and form copper oleates, acetates, palmates, etc. These are generally soft, waxy, blue-green substances that generally do not cause disruption of the metal due to their softness.

The corrosion products of both iron and copper are often larger than the metal from which they came, and rise above the original surface. Wherever wood and metal are closely joined, expanding corrosion products compress the wood, causing splitting or similar damage. At the same time, dissolution of the cellulose by metal ions in solution can collapse the cell structure of the wood. Staining of the wood caused by dissolved metallic salts is also very common. If relatively fresh wood is combined with iron, further drying can cause cracking as the metal resists the movement of the wood.

Up

Stable Composite Objects

• The metal has developed a thin passivating layer that prevents further corrosion from taking place. This can occur on both iron and non-ferrous metals.

• The moisture content of the wood is low enough that corrosion reactions are not initiated.

• The wood contains extractives like oils and waxes that retard corrosion or provide a physical barrier.

In many cases, a combination of all three mechanisms is responsible for stability.

Degraded composite objects can be very fragile because of damage due to corrosion products, and weakening of the wood cell structure due to metal salts. High relative humidity may accelerate corrosion, while low relative humidity will affect wood. Stabilizing composites of metals and wood presents major problems where active corrosion is taking place, and where some form of environmental control is considered necessary. A dry environment is preferred for stabilizing metals. However, dry environments for wood usually result in cracking, warping and other distortions. Obviously, a compromise is necessary in such cases. In general, a relative humidity of around 35% will stabilize most corrosion products, while not subjecting the wood to extreme conditions.

Up

Wood and Other Substances

Chapter Head

Return to Contents

Treatment is approached in the following sections in two ways: by material of fabrication, and then by type of instrument. The first section generally describes applicable techniques for a wide range of materials and conditions, and the second deals with the problems of specific musical instruments.

Basic advice on the care of displayed musical instruments is provided by the National Trust Manual of Housekeeping and similar texts, but the following guidelines address treatment needs that go beyond routine "housekeeping" duties. It is assumed that once any of these techniques has been performed on an instrument, ongoing care will be exercised to keep it in good condition. It is pointless to clean and stabilize instruments if they are then returned to the conditions that contributed to their state.

The techniques described here require some simple equipment and a knowledge of its use, and a few materials and supplies. Commercial products should be used with some caution because their ingredients are often not known and manufacturers can change formulations without notice. Recommended products are noted in the text. Some familiarity with musical instrument terminology is assumed.

The conservation profession is widespread and accessible, so if there is any doubt about treating an instrument, resources are available. This is especially important when suggestions or recommendations for treatments have been made that go beyond simple stabilization and presentation. The resource list at the end of the book includes the addresses of some national conservation associations, advisory bodies, museums and funding agencies.

Chapter Head

Justifying Treatment

• What caused the deterioration?

• What techniques have been used in its construction?

• Is it stable?

• What did it look like originally?

• Is its appearance acceptable now?

• What do we want it to look like?

• What do we hope to gain by treating it?

Articulating these questions develops a justification for action. Because of the nature of musical instruments, the variety of possible fabrication techniques, and the range of possible conditions, it is essential to learn as much about the object as possible.

• to make it look better?
• to make it work better?
• to stabilize it?
• to gain information?

All the above six categories suggest further questions, none of which can be answered fully here. The questions serve two purposes. First, by asking questions we initially put the brakes on our hands. This is one of the less understood functions of documentation ­ it obliges us to rationalize our approach and create a logical and answerable sequence of steps. Secondly, by asking questions, we gain a better understanding of the material under our care and our approach to it. Inevitably, though, no matter how conscious one is of the above questions, it is difficult to provide all-encompassing guidelines; each instrument raises specific problems.

Chapter Head

Review of Cleaning

Definition of Surface

The attractiveness of a surface to dust, dirt and accretions with use depends upon its nature. For example, a rough or fibrous surface may not attract airborne dust more readily, but it will help it stay in place. On the other hand, the same rough or fibrous surface will cause accretions of use to accumulate faster than they would on a smooth surface.

Up

Definition of Dirt

A dirty surface can actually be a very complicated phenomenon because the dirt may consist of foreign material, chemically combined material, or a mixture of both.

• Foreign matter. This is material that does not originate from the instrument, but which has settled on it. Examples are dust blown in from outdoors, and grease or dirt caused by handling.

• Product of alteration. This is material that originates from the instrument, and which has then been transformed by chemical reaction with its environment. The most obvious example is corrosion products on metals, but many other substances, such as wood and leather, oxidize on exposure to the atmosphere.

• Physical mixture. This is a mixture of material from the instrument and foreign matter. Any porous or fibrous surface can absorb dirt in solution in water, which then dries out leaving the dirt and the object indistinguishable. Tide lines on wood that has been wetted and dried are physical mixtures of dissolved dirt and wood fibres.

Up

Why Dirt Should be Removed

• Abrasion. Some dusts, especially those carried into a building during construction work, can be very abrasive. Handling dusty objects can scratch the surface, and might embed granules in the object. This is especially important with varnished wooden instruments or highly polished metal ones.

• Attraction of pests and mould. Organic dust, found most often in the inaccessible interiors of instruments, provides nutrients for many organisms including fungi and insects.

• Sites for reaction. Dust provides a focus for chemical reactions, particularly corrosion on metals. When relative humidity rises dust can attract moisture, thus stimulating oxidation reactions.

• Chemical reaction. Some forms of dirt chemically react with the surface, and thus change its nature.

• Masking. Dirt and dust can mask a surface, obscuring details and giving the object an unpleasant appearance.

Up

Why Dirt Should Stay in Place

• Potential for damage. Removing dirt might be so difficult that it causes damage to the surface of the instrument. From a conservation point of view, this is the chief reason to be cautious when dealing with dirty surfaces.

• Protection. A layer of dirt may be protecting a surface, and its removal might cause a further undesirable reaction between the surface and its environment.

• Information. The dirt may contain information useful to the interpretation of the instrument or its context.

• Aesthetics. Patterns of dirt may contribute to the appearance of an old instrument.

If dirt were simply to settle on an object by gravity, it is reasonable to assume that it could be easily removed by turning the object upside down. Obviously, this does not happen, so the dirt must bond to the surface in some way. In fact, there are several ways in which dirt can stay in place. These are, in order of removability, mechanical attraction, electrostatic attraction and chemical bonding.

Up

General Principles of Cleaning

It should be decided if cleaning is truly needed and how far it should go. What is "museum dirt" and what is "dirt from use"? Can one be removed and not the other? As a rule of thumb, more recent dirt from handling and storage in inadequate conditions will be easier to remove than dirt accumulated on an instrument from use. Learning to distinguish between the two, and knowing where one stops and the other begins is acquired by experience, and is not easy to teach.

Up

Review of Techniques

• Mechanical cleaning. This includes any method that relies on the physical action of one object upon another. It includes blowing, dusting, rubbing, erasing, abrading and chipping, and can be done with the range of tools from soft brushes to hammers and chisels. Mechanical cleaning works best when the dirt and the surface are well defined; in other words, where the dirt and the surface are easily separated without damage.

• Solvent cleaning. Solvent cleaning involves the use of any fluid that has the capacity to dissolve or loosen dirt from a surface. There are a variety of mechanisms involved in solvent cleaning including:

• Dissolution, where dirt is dissolved in a solvent and washed off the surface.

• Softening, where dirt is softened and can be removed mechanically.

• Chemical cleaning. This involves chemical reaction (oxidation or reduction) that occurs between the surface of the object and a cleaning medium. The methods include:

• Enzyme reaction, where an enzyme is used to catalyse a reaction. Such techniques are most often used in cleaning organic residues from paper and textiles.

• Sequestering agents, where molecules cluster around positive or negative ions causing them to become soluble.

• Electrochemical and electrolytic reactions, which are especially used in cleaning metals.

Chapter Head

Treatment of Wood

Cleaning Surfaces

• soft brush and vacuum cleaner

• stiff brush

• eraser powder

• adhesive rubber

• gum eraser

• damp swab

Up

Soft Brush and Vacuum Cleaner

Figure 14. Brushing dust into a vacuum cleaner nozzle.

The nozzle should be covered with cheesecloth held in place with a rubber band to prevent small pieces being accidentally sucked up. A speed controller is useful for regulating suction (see below). Vacuum cleaning should not be done if loose pieces, powdery surface coatings or delicate components are present. Soft brushing without the vacuum cleaner is possible if the surface will permit it.

For access to intricate areas (i.e., underneath the strings or between the keys of a keyboard instrument), a narrow hose attachment can be made for the nozzle of a vacuum cleaner. A piece of 4mm bore flexible plastic tubing can be attached to the nozzle with tape.

Figure 15. Hose attachment for a vacuum cleaner nozzle.

Because the narrow tube constricts air flow, the vacuum cleaner motor may overheat.

This can be relieved by providing a pressure relief hole in the nozzle, which is covered by a finger when in use, or by connecting a speed control into the vacuum cleaner circuit. If the latter technique is chosen, an electrician should be consulted, because certain kinds of alterna-ting current electric motors cannot be controlled without damaging them.

A miniature vacuum cleaner of the type used in the electronics industry is useful for the finer applications in conservation, or one can be constructed by reversing the flow of an aquarium pump (CCI Notes 18/2).

Up

Stiff Brush

Figure 16. Microscopic cross section of wood, showing the effects of wear due to abrasive cleaning, which has enhanced the difference between early- and latewood cells.

Up

Eraser Powder

Up

Adhesive Rubber

Up

Gum Eraser

Up

Damp Cleaning

Up

Cleaning Cracks

Up

Repair

Occasionally traces of old glue provide a "key" to the original position of detached components, so it is sometimes advantageous to leave these traces in place, because they allow precise repositioning of parts in their original location. This is especially applicable if the glue is not on structural components.

Readhesion should always be done with a natural adhesive, not a synthetic modern product. Adhesives such as the commonly-used white glues are very convenient, but are difficult to re-dissolve. Using natural glue will ensure that the repair can be taken down in the future if necessary. Adhesives traditionally used for fine woodwork were hide glues, applied while hot and allowed to cool in place. The techniques for hot gluing are hard to master, so for the occasional repair by the non-specialist, it is better to use a liquid hide glue, or a fish glue. These can be injected or brushed into the joint, slightly thinned with water if necessary. A hypodermic syringe is the best tool for putting the glue exactly where needed, and nowhere else. Only the minimum amount necessary for the join should be used. Once applied, the join should be compressed briefly by hand to remove excess glue. Any glue that oozes out should be wiped off before clamping.

The secret of a good glue join is in the closeness of the parts, which is achieved by efficient clamping. Natural glues are poor gap fillers, so the less glue there is in the join, the stronger will be the finished result. A wide range of clamps is available from specialist woodworking suppliers, the size and type of clamp depending upon the work to be done. Clamps are never brought directly to bear on old surfaces; there must always be some form of padding between the jaws and the work. Also, silicone paper, wax paper or thin plastic film is applied so that any excess glue that may ooze out will not stick to the clamp or pad. Small pieces of 2.5mm thick Plexiglas (Perspex) with smooth, bevelled edges are useful as clamping pads. It is advisable to do a "dry run" by attach-ing all the clamps and pressure pads, without applying the glue, to check the security of the join. However, this may not be possible or advisable when the pieces involved are too small or vulnerable to breakage themselves.

Repairing damages by gluing is a relatively complicated process. It is far better not to do it than risk making things worse by doing it poorly. If the equipment is at hand, it is advisable to practice on a few worthless pieces of wood first to get the feel for all the things that can go wrong.

Chapter Head

Treatment of Metal

Pros and Cons of Polishing

Figure 17. Microscopic cross section of a metal surface showing how polishing to remove tarnish always removes metal as well.

Commercial metal polishes vary in their abrasive properties from the very mild to the extremely coarse. Only a few polishes are considered useful for historic material, and even these must be applied with great caution. Because metal is always removed from the surface, the treated piece becomes thinner and decorative detail is worn away. Also, there is no point in polishing the metal parts of an instrument if it is destined to return to its corroded or tarnished condition fairly quickly. Polishing entails the responsibility, first, of removing original material forever and, second, of maintaining the newly exposed surface.

Metals occasionally have an artificial, chemically applied patination that alters the appearance of the surface. The solid black finish of some 19th century brass instruments has been confused with tarnish and has been incautiously treated.

Up

Basic Treatments

• Cleaning. This can be achieved mechanically, chemically, electrolytically, or electrochemically. For the purposes of these notes, mechanical cleaning in the form of abrasive removal is the primary technique recommended. Products available for this include cloths, pastes, waddings, and fluids.

• Stabilization. This may be done with or without corrosion products in place. It can involve removing harmful material from the metal, or applying a protective layer to the surface that lowers its reactivity. The protective layer is chemically bonded to the metal and should not be confused with the physical barrier provided by coatings.

• Coating. This is the application of a surface layer (lacquer, wax, or paint) in order to isolate the metal from any hostile aspects of its environment. A coating can protect against pollutants, dust, and the residues from handling, but is never a complete barrier. Small flaws or breaks in the coating can actually induce pitting corrosion.

Up

Cleaning Iron and Steel

Commercial rust strippers are not recommended. They are often based on acids (particularly phosphoric) and may contain a number of unknown additives. Formulations may change without notice, and it may be difficult to remove residues.

Tannic acid can be applied on an iron surface in order to form iron tannate (a dark blue-black colour), which is stable and acts as a slightly protective layer. This should only be used on forged or cast iron parts whose surface was originally black (CCI Notes 9/5, 9/6 and 9/8).

Many steel and iron parts were originally patinated by either heat or chemical treatment. Most tuning pins have an artificially applied black or brown patination, which should not be removed.

Up

Cleaning Copper and its Alloys

Up

Cleaning Silver

Silver dips can be used on parts of instruments that are detached and do not contain hollows or other attached metals. Great care must be taken because silver dip is not discriminating; it will remove all tarnish, giving a very flat appearance to the surface. This is especially important on pieces with etched, engraved or incised decoration. A dip which has been used on any other metal should not be applied to silver because baser metals held in solution will "plate out" onto the silver.

Coatings for silver, and the copper alloys mentioned above, include lacquers and waxes. Generally, lacquer coating is not recommended because it is difficult to remove it later. A thin coating of microcrystalline wax, such as Renaissance Wax, is more innocuous; it can be removed with mineral spirits and reapplied as needed (CCI Notes 9/7).

Up

Cleaning Lead and its Alloys

Wax is the most commonly used coating for lead objects although, given adequate storage or display conditions, it is unnecessary. The chief treatment for lead is passive ­ providing apposite environmental conditions.

Up

Repair

Chapter Head

Treatment of Other Materials

Cleaning Skin and Leather

• rawhide and parchment most sensitive

• alum tawed skin

• semi-tanned skin

• semi-tanned and smoked skin

• fat liquored (chamois) skin

• vegetable-tanned leather

• mineral-tanned leather least sensitive

Cleaning treatments in order of increasing aggressiveness are:

• soft brush and vacuum cleaner

• stiff brush

• saddle soap

• eraser powder

• adhesive rubber

• gum eraser

• damp swab

• wet swab

Up

Cleaning Rawhide and Parchment

Up

Cleaning Leather

Up

Softening Rawhide and Parchment

Up

Softening Leather

In one exceptional case saddle soap may be used: the smooth (grain) side of vegetable-tanned leather, if it is in sound condition with no surface cracks or defects, may be polished. Saddle soap is applied sparingly with a soft cloth and buffed to a shine. This acts as a cleaner and also gives a fine protective coating to the leather. Saddle soap is alkaline ­ it contains saponified fatty acids (triglycerides) ­ while the natural pH of leather is around 4 (acidic). There has been some debate on the efficacy of applying an alkaline material to a naturally acidic one, thus neutralizing it or, at least, somewhat raising its pH. So far no detrimental effects have been noted.

Up

Cleaning Bone, Ivory, and Horn

• soft brush and vacuum cleaner

• stiff brush

• eraser powder

• adhesive rubber

• gum eraser

• damp swab

• ethanol/water mixture

• ethanol

• mineral spirit

• acetone

Only under extreme conditions will it be necessary to use strong solvents. Normal "museum dirt" can usually be removed mechanically with less aggressive methods. Deposits of dirt in cracks in ivory can be removed carefully with sharpened wooden picks. Metal tools should not be used. Caution in cleaning highly carved and decorated ivory is recommended because removing all accretions in the lower relief of the design can leave a bland and uninteresting appearance.

Yellowing of ivory (key facings, wind instruments, etc.) is usually not reversible by any cleaning method. The traditional methods of metal polish, toothpaste, and bleach cannot be recommended. Exposure to ultraviolet light causes a bleaching effect on ivory, thus it is occasionally recommended that keyboards be exposed to sunlight to keep the ivories white. This is not a desirable treatment for historic objects (CCI Notes 6/1).

Up

Repair of Bone, Ivory, and Horn

Clean, well-aligned breaks may be glued with a PVA emulsion white glue. This glue yellows with age, so it is advisable, first, to make the join as close as possible using the minimum of adhesive and, secondly, not to spread the glue right to the outer edges of the break. Thus, when yellowing takes place the join will still be invisible. PVA emulsions require overnight drying when used on less permeable materials because they dry by evaporation. Hide glue has also been used successfully for repairs to bone and ivory. Repairs made with these adhesives cannot be expected to be very strong. Stronger bonds might be possible with polyester and epoxy resins, but they are almost impossible to unglue at a future time.

Horn is not as easy to glue. Its surface resists the penetration of most synthetic resins, which then tend to peel off. However, damages to horn will not be encountered as often because it is more flexible and thus is less likely to crack.

Chapter Head

Treatment of Synthetic Organic Materials

Deterioration of ebonite is accompanied by an exudation of acidic material. If this is suspected, a simple test with pH paper will confirm the presence of acidic materials. In such cases the advice of a conservator should be sought (Bacon).

Chapter Head

Cleaning of Textiles

Soft, open felts are particularly prone to accumulating larger sized particles of dust, insect casings, and other detritus. This is particularly noticeable beneath keyboards. Large, superficial material should first be removed from the surface using a pair of fine forceps. Once the larger particles are removed, a piece of cheesecloth should be spread over the whole textile and held down firmly by placing weights at its edges. The piece can then be vacuum cleaned through the cheesecloth using a speed controller as described earlier. If this level of cleaning proves insufficient it will be necessary to consult a textiles conservator (CCI Notes 13/7).

Chapter Head

Treatment of Coatings

Up

Pigment

Consolidating fragile surfaces is often necessary, but it is rarely accomplished without some change to the surface appearance. If consolidation is considered essential, the change in the nature of the surface must be thoroughly discussed and a true understanding of the projected treatment gained by tests or research.

Up

Paint

The composition and deterioration of painted surfaces is a very detailed study, and all work done on loose and flaking paint should be performed by a specialist. Flaking paint surfaces are consolidated by running a suitable consolidant in solution or emulsion under the flakes. Where distortion of the paint film has taken place (e.g., the flakes are curved upwards, or cupped), heat and pressure may need to be applied. Occasionally, the solvent used for the consolidant will soften the paint, allowing capillary action to flatten the flakes. Losses in painted surfaces are sometimes inpainted, or retouched, to preserve the overall appearance of the object.

Up

Lacquer or Varnish

Lacquers and varnishes can be readhered to their substrates under some circumstances, but success depends upon the relative transparency of the solution used in consolidation, its refractive index compared with that of the varnish film, and the effectiveness of penetration below the loose flakes. Transparent coatings can become more obtrusive after consolidation. In some circumstances lacquers and varnishes that are clearly not original are removed during treatment and reapplied.

An intact lacquer or varnish that is dull, but which shows no cracks or gaps in its film, can benefit from wax polishing. The best choice is a white paste wax with no added colorant or perfume, and especially without silicone. Silicones are added to many commercial paste waxes to give a high shine and resistance to moisture, but their disadvantage is that once applied they are difficult to remove later. Plain white paste wax can be applied thoroughly to the finished surface with a soft cloth and allowed to dry to a dull film before buffing to a shine. Build-up of excess wax around attached hardware should be avoided. It is not necessary to wax polish at frequent intervals unless the instrument is handled excessively.

Up

Oil or Wax

Although not a surface treatment, it is worth mentioning that wax was occasionally used as an aid in tuning by using it to adjust the size of woodwind holes. Also, instruments made to be played either left or right handed may sometimes have one of the lower holes for the little finger filled with wax. Traces of such wax found in woodwind holes should remain undisturbed.

Up

Veneer

Loose inlays and veneers are relatively easily readhered provided there is no distortion. However, the reason the pieces have become detached is usually due to the substrate having moved, often in response to changes in relative humidity. In this case, careful reshaping or complete replacement are sometimes necessary. Metal inlays are more problematic to readhere due to the difficulty of finding an adhesive which is strong enough, yet still removable. Traditionally, bone or hide glues were used, and although not strong they still have the virtue of removability. Most work on veneers and inlays is specialized in nature.

Up

Enamel or Glaze

Flaking enamels can be readhered in some circumstances, but specialist assistance should be sought. Crazing on ceramics and similar substances is rarely treated.

Up

Plating

Gold or silver leaf applied to non-metal surfaces must be treated extremely cautiously. If the surface is stable and intact, light cleaning with a soft brush can be done with safety. Water should never be al-lowed to come in contact with gilded surfaces. If there is any evidence of cracking or loosening of such finishes, specialized help should be sought.

Up

Artificial Patination

Patinated metals must be treated with extreme caution, especially in places where an artificially oxidized surface may give clues to heat treatment of parts. Polishing of such surfaces is not recommended.

Chapter Head

Treatment of Composites

If the parts can separated, they should be treated individually using appropriate techniques for each material. It is generally advisable to remove all corrosion products from metal parts before reassembling a composite object because corrosion is often disruptive, and can be reactivated in the future. Usually such corrosion products occur where metal and wood are in close contact, so their removal is often not visible on the reassembled object. Chemical treatments such as stripping, which are normally not recommended for visible surfaces, may be used in this case. Distortion of wood by corrosion products often means that repair of the wood is necessary before reassembly. The alignment of both metal and wood parts should be checked before any glue joins are made. A protective layer such as wax or lacquer should be applied to the metal before reassembly.

Where parts cannot be separated, it is only possible to treat the visible surfaces. The metal must be treated by whatever method is found appropriate, but the wood must be well protected from contact with chemicals and materials used when treating metals. Oil and tannic acid solution will leave permanent stains on wood, and marks can be left by metal polishes and carelessly applied mechanical methods. In general, treatments for wood will have no effect on the metal, so covering and masking is often not necessary when working on wooden components. Treat-ment documentation should make it clear that full treatment is not possible while the parts are immovably joined.

Iron stains on wood are a common occurrence. They are usually orange or light brown and disfiguring, and can cause decay of the wood. Stain removal techniques usually result in further softening and degradation. If cleaning and stain removal have been done, wood parts may need consolidation to strengthen them.

Stains from copper salts are green or blue and tend to be more difficult to remove. Complexing agents such as EDTA are sometimes successful in removing stains but, as with iron, the wood may be further damaged as a result. Corrosion products from the reaction of copper with organic materials can usually be removed mechanically, followed by swabbing with acetone or turpentine substitute to further clean the surface. Solvents should be used sparingly on wood because substances in can be carried further into the wood by capillary action.

Chapter Head

Treatment of Instruments by Type

Up

Keyboards

The process of slackening the strings should be undertaken over a period of weeks, to avoid shocking the structure by suddenly relieving it of tension. Using a well-fitting tuning key, begin detensioning in the centre of the compass and work alternately to the bass and the treble by octaves in the pattern of setting a temperament (i.e., by fifths and fourths). Unlike normal tuning, all sets of strings should be detensioned together, rather than consecutively. In other words, do not detension one eight-foot register of a harpsichord before beginning to detension the other eight-foot, or the four-foot. Because every string cannot be detensioned simultaneously, the pitch on any one string should not be lowered by more than a semitone at a time, otherwise there is a risk of breaking those strings not yet slackened. Detuning every string by semitone increments can be done on a daily basis. Iron-framed instruments can stand a much more accelerated schedule of detensioning. On completion, the strings should have just enough tension to lie straight, and remain on their tuning pins.

Exterior wood finishes that are in sound condition can be wax polished if necessary, but such work is time-consuming if done properly, and the results may not warrant the effort. Small tests of wax polishing can be made to determine the degree of change in appearance and time taken to achieve the result.

The chief collecting places for dust are between and under the keys, and on the soundboard under the strings. Dust can be lifted with a small brush while the nozzle of a vacuum cleaner is held close by. Alternately, two vacuum cleaners can be used, one set to blow the dust away and the other to suck it up. A fine jet of air can be directed from the blowing vacuum cleaner by constricting the opening of the pipe, but this may cause overheating of the vacuum cleaner if done to excess. Caution should be taken when working beneath the keys because textile padding materials may be quite degraded and fragile. Also, the lifting action of the suction can fluff felts up, thus changing the regulation of instruments in playing condition. Painted decorations on soundboards should also be treated with caution because they are in a protected area, and are sometimes more fragile than they appear (McTaggart). Brushes with metal ferrules should be avoided, especially when working between the strings, and other places where the brush fits tightly and could cause scratching. Brushes with completely wooden shafts, like those used in Japanese paper conservation, can be used, or the metal parts of brushes can be wrapped with a soft material such as plastic sleeving.

Up

String Instruments

The loose ends of metal strings projecting from hitch pins, tuning pegs, etc. can abrade finishes. It is not necessary to trim the ends of the strings, and this should certainly not be done if they are old. String ends should be bent away from contact with delicate surfaces. It is especially important to prevent wire string ends from damaging the instrument when strings are being removed, replaced, or tuned. A little adhesive tape applied temporarily to the ends will help.

Before adjusting the tension of strings, the tuning pegs or machine heads should be carefully examined. All mechanical parts should have the potential to move smoothly before any effort is made to turn them. A small amount of light lubricating oil ­ enough to coat without excess ­ can be applied to the bearing surfaces of metal parts. Tapered wood pegs sometimes seize in their sockets. Abrasion of the wood, and even cracking of the pegbox, can occur if they are turned incautiously without first checking their freedom of movement. If a peg appears to be seized or moves stiffly, sideways pressure should never be applied. The peg may be gently turned back and forth with a pulling motion so as to release it. Be careful that the string wrapped around the peg does not find its way between the peg and the socket. If gentle pressure fails to move the peg, the advice of a specialist should be sought. Once wood-to-wood components have been separated, their contact surfaces can be rubbed with a thin layer of soft soapflakes to which a small amount of chalk dust has been added. This will avoid seizing in future while still providing grip. Hill's Peg Compound is recommended.

Interior cleaning of instruments with accessible soundboxes is recommended if they have been kept in dirty conditions or if loose dirt is visible. An air jet from a small piece of soft, flexible plastic tubing can be directed into the interior while the dislodged dust is collected by a vacuum cleaner nozzle. This must be done with care, and the approach will depend upon the instrument. Potentially fragile items like maker's labels and parchment linings should be dealt with cautiously. It is not necessary to clean interiors regularly.

Up

Woodwind Instruments

Metal keywork on wooden instruments sometimes suffers from corrosion due to contact with oils and acids in skin, lubricants, or cleaning substances used on the wood. This corrosion is usually fairly soft and can be removed mechanically with a sharpened wood stick, followed by gentle swabbing with turpentine substitute or a similar mild, petroleum-based solvent. Wood surfaces must be protected from contact with the solvent by placing a small square of Mylar or polyethylene under the key being worked upon.

Keywork should be free-moving and in good mechanical condition, even in static instruments, because there is always the potential for damage during examination. All pivots should be checked for smoothness of operation, and the existence and condition of springs verified. A small quantity of light lubricating oil may be applied to the pivots and joints of metal instruments, but no lubricant should be applied to mechanical components of wooden instruments unless done extremely cautiously. Lubricating oils are difficult to remove from wood and can cause permanent staining. The keywork on old instruments can be in a very delicate condition, and nothing beyond light maintenance should be attempted. Under no conditions should mechanical work be done on keywork of wood-bodied instruments.

Padding on keywork should be treated very cautiously. The traditional adhesives, shellac and sealing wax, may be used sparingly when readhering loose pads. Shellac may either be dissolved in alcohol or melted by applying gentle heat to the key. Modern stick adhesives, such as those used in electrically heated glue pistols, may also be used. The choice of adhesive may be influenced by cosmetic considerations. Any adhesive appearing at the edge of the pad needs to be as invisible as possible. Adhesive should be used sparingly to form the weakest, most easily removable bond possible. Padding should never be removed unless it is unequivocally not original, and is completely useless or deleterious. Any original materials that must be replaced should be saved.

Joints of instruments should be checked for secureness. Loose joints can fall apart unexpectedly, while tight joints can contribute to further damage. Joints may be loose because thread bindings have deteriorated or have been improperly applied. In general, it is better not to rebind joints, but to store and display the instrument in such a way that its weakness is accommodated, and known to those who handle it. The cross sections of old wooden instruments may have become elliptical due to differential shrinkage of the wood. Under these circumstances it may not be feasible to return the joints to their original state.

Where joints between wood components are seized, or move stiffly, extreme care is needed. Any force applied to the parts must be rotational and longitudinal only; a combined pulling and turning effect rather than separate, rectilinear twisting and pulling. The pieces should never be rocked sideways in relation to each other because this places strain on the joint and will cause the wood to crack. It is common for thread bindings to loosen from a tenon in a manner that makes it difficult, if not impossible, to separate the joint. Stiff joints can be eased by wrapping one hand on either side of the joint, pressing the hands closely together, setting the line of their contact at a slight angle to the axis of the bore, and then rotating the one hand against the other. However, it may be unlikely that the joint can be rotated more than a few degrees in this manner due either to design constraints (keywork and other features) or to the joint having become radially deformed over time. Therefore, the joint must be returned to the starting position, but without losing the extension. This process may need to be repeated many times, and it will often be necessary to use a pair of fine forceps to pick out little bits of the binding as it becomes visible through the slowly widening gap.

Figure 18. The correct way to disassemble an instrument at the joints. The rotary and longitudinal forces are applied incrementally. There is no rocking movement between components.

Metal, horn or ivory reinforcement rings are often placed over the socket ends of woodwind joints. Shrinkage of the underlying wood can loosen these pieces. In order to prevent them moving, it is often necessary to bulk out the space between the ring and the wood with strips of very thin paper. The thickness of the the paper should be chosen carefully so that the pieces do not fit tightly together.

Up

Brass Instruments

Parts of metal instruments that have seized together are the most problematic to treat. If moderate force does not move them, there are three possible recourses: lubrication, heating, and ultrasonic treatment. A very fluid lubricant ­ a light machine oil in a volatile solvent base, known as penetrating oil ­ can be applied around the component and allowed to penetrate the join by capillary action. A dam of soft paraffin wax can sometimes be built around the joint to form a pool and encourage the oil to flow where it is needed. Once penetration is judged to have been effective, moderate force is applied again. Slight warming with a stream of hot air can assist penetration, but this should only be done if there is no surface finish such as lacquer, paint or patination on the instrument. Open flames should never be used. Ultrasonic treatment is often effective, but it requires expensive equipment and should be referred to a specialist.

Residues of old abrasive polishes are often found on brass instruments, especially in intricate areas of design or inaccessible places. These deposits are usually powdery and are often light green in colour due to reaction with copper. It is possible to remove this material with a sharpened wood stick, although it is tedious and time-consuming. The deposits can be moistened with a little distilled water to soften them, and the area dried once the deposit is removed.

Full-scale polishing of old brass instruments rarely produces aesthetically acceptable results, but if this approach is chosen one should try to achieve an overall balance in appearance. This means choosing a milder polishing treatment and working lightly over the whole surface, slowly bringing all parts to an even appearance. Chemical methods of cleaning non-ferrous metals have been used, but their control is difficult and they should only be applied by a specialist.

The interiors of old brass instruments usually hide extensive corrosion. Although corrosion is quite often passive, there are occasions when it can become active. Disassembly of valves, tuning slides, and other detachable components will help when examining the interior. The only truly effective way of assessing conditions in long and convoluted tubing is with a bore scope. This is a fibre-optic viewing device with a built-in light source that can be inserted to a considerable depth. If traces of brightly coloured corrosion are visible, it will be necessary to remove and stabilize it. Nylon bristle cleaning brushes for use on brass instruments should be used to remove loose deposits. Selecting the correct brush size for the bore is important. Once as much loose material is removed as possible, the affected area should be coated with wax to isolate and stabilize it. A soft, white paste wax, such as Renaissance Wax, should be diluted to a creamy consistency with a petroleum-based solvent, and then applied by pulling a piece of folded cloth soaked in the mixture through the bore on a string. The cloth must fit smoothly, but not tightly, into the bore. This treatment is the most expedient under the circumstances, but it may not provide complete stabilization. A conservator should be consulted if the corrosion appears heavy or unstable.

It has been reported that the playing quality of some brass instruments deteriorates markedly after interior cleaning. It is argued that deposits of passive corrosion in the interior, built up during the playing life of the instrument, contribute in some way to idiosyncratic tone formation. While this contention has never been tested systematically, it is mentioned here as an advisory.

Up

Percussion Instruments

Chapter Head

Return to Contents

Using any object causes wear and degradation, which implies that routine maintenance and repair, and replacement of worn components will be necessary. Wear with use may be obvious and require continual attention, or it may be difficult to detect over a period of days, weeks or months, and only become significant, harmful and obvious over a period of years. This is a well-known problem when exhibiting light-sensitive collections such as textiles and watercolours, and analogous problems occur when functional objects are repeatedly handled, operated, and maintained. As outlined in Chapter 1, if using historic instruments is a policy of the individual or collection, hard decisions must be made. The few factors that may weigh in favour of a functional restoration have been discussed, including the proviso that "the instrument has been previously restored and most ephemeral features have already been lost." The conditions that must be met under these circumstances, in addition to those cited earlier, are that:

• the instrument is already in working condition and its continuing use will not detract from its technical and cultural value;

• the instrument is in a stable and safe condition;

• qualified expertise is available for regular tuning, maintenance, consultation and monitoring;

• the owner or custodian has a clearly articulated policy on the instrument's use, which meets the standards of ethics and practice of an appropriate national conservation or museum organization;

• the instrument is not to be treated as a working tool in the everyday world, but that its historic value is understood and appreciated; and

• every effort is made to make players and audiences aware of the speculative nature of the instrument's "original sound."

Only when all these conditions are fulfilled can the use of historic musical instruments be considered. Even then, the future place of the instrument in the collection, and its long-term care and maintenance, must still be causes for concern.

Concepts such as "acceptable risk" and "historic value" when applied to projected restoration projects have such large subjective components that they become virtually meaningless when formulating the general guidelines presented in this book. Many musical instruments have historic value that is beyond all question; for example a single known violin labelled with the verifiable date 1526. Others, such as one example taken directly from a 1996 factory run of 10,000 plastic clarinets, would be judged expendable. Most collections will contain material that is distributed along the full spectrum between these extremes. No two collections are likely to contain the same distribution and, consequently, no generally applicable conservation guidelines can be issued other than the above-stated, extremely important basic principles.

The value of a collection is often determined by the proportion of significant material that it houses. The value of the conservation approach is determined by its ability to consider every single object in a collection on its own terms. Although flexibility is a requisite aspect of the conservator's everyday practice, the bulk of writing on the subject of conservation addresses situations where a dogmatic stance is necessary or, at the very least, defensible. The authors of this book have often gone on record as regarding many of the practices described here as being unacceptable under any circumstances. Such inflexibility is often justified because it can never be assumed that the individual responsible for treating an object can judge when deviation from a "hard line" might be permissible; all the more so when the objects of the treatment can be both invaluable and wholly irreplaceable.

This chapter on maintenance is included for several reasons. In the first place, it has been written with a clear statement of warning to the reader: the description of a procedure does not provide any a priori indication of when its use is called for. More importantly, it does not confer the ability to carry it out. Every word contained here is intended to provide the reader with an adequate understanding of selected conservation problems to enable a meaningful discussion with competent specialists about the scope of the assignments to be given to them.

A second, significant, reason for the change in attitude is the continuous expansion of the period with which the historically oriented conservator must deal. Period material now includes virtually all instruments not in current production. A vintage instrument from the 1930s is as likely to be entrusted to a museum conservator as to a commercial repairperson. Similar conditions apply, for example, to synthesizers that are only a few years old, no longer in production, and clearly significant to the history of electronic instruments. It is no longer possible to define the importance of an instrument simply in terms of its age. As a corollary to this, it is not possible to exclude modern workshop practices from the conservator's immediate concern.

Finally, even the most dogmatic of the authors of this book knows full well that the objects of their attention have been "restored to playability" even when the risks were of staggering dimensions, and that the practice continues. The following notes are therefore intended to alert the reader to the complexities inherent in any decision to continue playing status, and to emphasize the need for qualified care of such instruments. To provide detailed instructions on how to regulate complex historical instruments such as harpsichords, pianos, and bowed strings would be inappropriate here. The required skills depend upon a combination of formal education and apprenticeship, plus the judgement gained by years of practical experience and specialization in treating a particular type of instrument.

Chapter Head

Keyboard Instruments

Up

Environmental Considerations

Up

Structural Concerns

Structural distortion shows in tensioned keyboard instruments in predictable ways. The pull of the strings, interacting with the structure of the instrument, generally causes not only a corkscrew deformation, but also a lifting and folding in of the tail and cheek. Such distortion puts many structural joints at risk. Among these are: the cheek-to-bentside joint, the wrestplank-to-cheek/spine joints, and the wrestplank-to-facia joint. The wrestplank itself may bow in the centre, or even become detached from its support blocks in the cheek and spine, causing the facia (or nameboard) to lift with it. Many 18th century English harpsichords and pianos provide startlingly severe examples of all these types of distortion. This underlines the previously stated caution, that seemingly infinitesimal short-term effects can become ruinous over longer periods.

Hitchpin rails show the strain of string tension where it is resisted nearer to 90^o. Thus, the extreme bass and treble hitchpin rails often bear the brunt of the tension, becoming either detached from the case sides, or worse, splitting along the line of the hitchpins. The process of separating from the case sides can be very slow, depending on environmental conditions, and action should be postponed as long as possible. Repair often involves replacing the compromised portion of the hitchpin rail, although consolidation treatments of the original material have also been performed successfully. Whatever course is taken, the original structure and components of the instrument are further compromised, and transfer of stresses from strong, new components may further weaken original material.

Soundboards respond very quickly to changes in environmental conditions because they are thin. In damp weather, boards are likely to dip or "trough." Equally, they may "crown" causing the four-foot bridges (together with their pins, of course) to rise and foul the eight-foot strings. Usually, faults associated with crowning and troughing disappear when conditions return to normal and no remedial action is required. In dry conditions, the soundboard will shrink and may split in the treble area, where it is held most rigidly. Hairline splits will probably close up again once the air becomes moister, but larger, unsightly splits may, for acoustical or cosmetic reasons, require attention. Successful repair of splits using long-grain wooden fillings, relies on the instrument being returned to very stable conditions. Otherwise, the areas of extraordinary strength along the filling's glue lines could cause the comparatively weaker board just beyond them to split when dry conditions arise again. In damp conditions when the board swells, the filling can also cause damage, because it will need to be accommodated in a space formerly occupied only by the old material.

The relative flexibility of the stringband (soundboard-bridges-string assembly) can cause seemingly unrelated problems with the action. In harpsichords, for example, a fall in the stringband can result in mild to severe damper inefficiency. This occurs because the dampers, mounted in the jacks as part of the action, rest ultimately on the robust case bottom, and remain comparatively stable in the vertical plane. The humidity-sensitive soundboard, on the other hand, can move down, taking the strings with it, out of contact with the dampers. Conversely, when the soundboard rises, the jacks may hang from the strings by their dampers. This begins to produce grooves in the dampers, making them incrementally less efficient in the event of a subsequent fall in the stringband.

Up

Exteriors

Inserting a prop or lid stick is no more straightforward. The soundboard should never support the weight of the lid. Instead, the bottom of the propstick should rest, depending on its shape, on the inside lower cheek moulding or on the case side (cheek) itself. The tip of the stick should be positioned in its lid housing or against the lid moulding. The security of the joint between the lid and its moulding should be checked periodically. A lid should never be left propped open for long periods.

Up

Harpsichords, Spinets, Virginals, and Clavichords

Clavichords possess a struck action, making them the true antecedents of the piano. Instruments in which one set of strings, struck at different points, serves as the sounding length for more than one note, are known as fretted clavichords. All others are said to be unfretted. Harpsichords, spinets and virginals all possess the same plucked action, with jacks, plectra and register(s).

Up

General Guidelines

When more than one set of jacks is engaged and a key is depressed very slowly, resisted or supported by another finger underneath the key, the jacks should pluck their respective strings not simultaneously, but consecutively, in a constant order throughout the compass. This non-simultaneity is known as the stagger. The distance between plucks should not vary appreciably on adjacent notes, though it may change gradually from bass to treble.

Up

Keyboards

Keys should respond quickly and quietly when played. Each key has a pin about which it pivots at the balance rail (the middle member of the keyframe), and one other guide point. Front guided keys normally have pins on the touch rail (the proximal member of the keyframe) over which the keys run. Back-guided keyboards may have a rack, a series of vertical wooden slots, in which metal pins or slips of smooth wood, which protrude from the distal ends of the keys, run. Other guide materials, though less common, include whalebone (baleen), ivory or metal sheet. Other back-guided methods include having each key travel between two pins or along one pin located in a channel in the distal end of the key. Such pins would be driven into the distal member of the keyframe.

When a key sticks or otherwise malfunctions, it is imperative that a specialist familiar with the action be called upon to establish where the fault lies. Treatment may involve removing the keyboard(s) from the instrument. Usually, keyframes simply rest inside the keywell, but sometimes they are fixed into place with turnbuckles, sliding pins or slats, or screws, and sometimes, as part of a restoration, the keyframe was screwed in place through the bottom boards of the instrument. Withdrawal of the keyframe is best done by a specialist.

A change in humidity frequently causes intermittent sticking of keys. In such cases, interventive treatment may be unnecessary because the fault will disappear when the humidity returns to normal. Sometimes, the key can be freed simply by rubbing it gently up and down over the pin on which it is binding, or by lightly burnishing surfaces where the key appears to catch. Dust in the rack slot or key mortise can also cause a key to stick.

Perversely, worn or enlarged balance (or other) mortises can also cause keys to stick. This type of fault very often appears when keylevers are cranked, by design, or run diagonally from one guide point to another. Such keys wear unevenly, and begin to rock, catching on the sides of the mortise or guide slot. Sometimes the effects of mild wear is ameliorated by attaching several layers of paper (for example, brown gummed paper 0.06mm thick) on either side of the mortise, on top and bottom if space permits, limiting excess diagonal travel and helping the key to run true. This is only a temporary measure, and eventually the mortise or slot will require repair as wear progresses. If either the mortises or the jacks are original, one should think hard about playing the instrument at all.

Up

Jacks

Historic jack design lacks nothing, and contains nothing superfluous. The adjustments devised by late 19th and 20th century restorers, however ingenious, never actually improved the performance of jacks. On the contrary, in this context they invariably act as hindrances. When original jacks have acquired adjustment screws, or have been replaced with over-engineered modern ones, the best course is to have them set correctly at the outset (at known temperature and relative humidity levels) and then forget them. Using them may cause real problems, developing elsewhere in the structure, to be ignored, and they may mask significant changes in the regulation.

In order to work well, jacks need to travel up and down freely in their register slots (see also Registers, below), but without excess side-to-side movement. The normal motion of the jacks wears them and their register slots, so that their fit may become too loose. Applying an appropriate number of layers of brown paper to the bearing surfaces of the jack body improves the fit, and is easily removable. However, thickening the jack body in this way will contribute to the inevitable enlargement of the register slot. If the instrument is to be used, such wear is preferable to the uneven wear and unreliable performance occasioned by loose-fitting jacks. Note that each jack is individually fitted to its register slot, and each plectrum voiced for its one string. Jacks are not interchangeable.

Up

Registers and Lower Guides

If movable registers begin to stick, this could indicate that, under the tension of the strings, the register gap is closing up. The reasons for such structural instability should be investigated. Among the questions to consider are: could the instrument be too heavily strung, are the gap spacers failing, is the relative humidity too high or fluctuating too much, or can the instrument still support strings at playing tension?

The overall movement of registers, from "on" to "off" position, usually needs to be no more than 1mm. If changes in humidity begin to make a whole register sound too loudly, the insertion of thin pieces of paper (about 0.03mm, the thickness of a cigarette paper) one at a time, in the "on" position of the register on the case side, can alleviate the temporary problem. (See also Machine Stops, below.)

Specialist advice should always be sought when leather-covered, or decorated registers, or those which form part of a machine stop mechanism, require modification.

Up

Buff Battens

The shape and size of the pads vary considerably, as does the manner of fixing them to their wooden batten, located directly behind the nut. This could be with an adhesive or with small nails or household pins. If detached pads are to be reattached with adhesive, always specify animal glue applied sparingly to the bottom surface and kept well away from the area of the pad which comes into contact with the string.

Up

Machine Stops

These devices, which vary in concept, design and mechanical realization, all facilitate quick changes of timbre or dynamic level by means of a pedal or knee lever. They may involve the withdrawal of one or more registers and often, the simultaneous engagement of another. With the machine stop engaged, players do not need to take their hands away from the keys to effect the change. It is imperative that players are instructed in the use of each machine stop. In particular, they must never attempt to move the handstops of the registers that form part of a machine stop when the machine stop is engaged. Doing so will disturb the regulation, and may damage the machine mechanism and other action parts. Where possible, machine stops should be disengaged after use.

Each machine stop should be considered individually, but the operation, engagement, and withdrawal of any of them should be quiet. Sometimes, a very slight depression of the pedal on engaging or withdrawing the machine stop facilitates a quieter operation and reduces wear and tear by helping to align metal leaf springs and hooks as they interlock or disengage. Advice should be sought before adjusting the end stops of registers which form part of a machine stop mechanism (see also Registers, above). Such alterations could affect more than one register and damage the machine mechanism itself. An unusual resistance in engaging the stop or depressing the pedal, or a squeaking or grinding sound associated with any function, indicates a fault which should be investigated by a specialist. Use should be discontinued until the fault is rectified.

Once set up, the mechanism itself should not be tampered with. Ease of adjustment and maintenance were not design priorities in typical English machine stops. Sometimes complete disassembly, including removing the registers, is required to make a small adjustment consisting of, perhaps, one turn of a threaded metal part.

Up

Tangents

Unfretted clavichords may be tuned, within the normal limitations, as the player wishes, but the layout of the tangents largely determines the temperament in which fretted clavichords must rest (see Action Types, above). Once the tangents in such instruments are fixed into place, there is very little latitude for modification. Evidence suggests that early makers of fretted clavichords bent tangents to achieve correct tuning, although it is risky to bend them as a measure of routine maintenance because brass can become very brittle with age. Also, in the absence of reliable information about specific examples, bending tangents to obtain varieties in tempering can be philosophically unsound. Ideally, temperament takes into account what the past placement of the tangents implies (see also Tuning below), but after restoration the present tuning may not faithfully reproduce the instrument's original temperament. If tangents work loose, specialist advice should certainly be sought.

Up

Listing

Up

Tuning

It should be noted that the techniques of early keyboard tuning, though beyond the scope of this book, differ substantially from those of the modern piano, and that proficiency at one by no means confers competence at the other. The tuning pins of early instruments are of no standard size or shape, and an ill-fitting tuning hammer can easily damage them. Well-fitting, custom-made tuning hammers are the choice of specialists. In addition, early keyboard instruments, unlike modern pianos, may rest at a wide variety of pitches and in a plethora of different temperaments, which vary according to the harmonic characteristics of the music being played. Among the other factors which may influence a performer's choice of temperament are the dates of the instrument and the music, and in fretted clavichords, the layout of the tangents. Early keyboard instrument tuners should not only be conversant with the different temperaments, but also with how to use them.

Up

Pre-modern Pianos

Action and Moving Parts

A high premium should be placed on the conservation of any piano action that retains original ephemeral or semi-ephemeral parts, such as beak leathers, hammer coverings, damper materials, kapsel stems and forks, bushings and so on. Many of the old materials are difficult to duplicate today, and all modern replacements present certain deficiencies. Our understanding of these actions has not yet reached maturity, and it is primarily through the study of those which remain relatively untouched, that comprehension can be advanced. Expert advice should be sought during the initial documentation of these instruments, whether or not original materials are suspected to be extant. Steps should be taken to discontinue use of any instrument that has original ephemeral parts.

There are many more moving parts in piano actions than in those of the harpsichord family, and far more significant variations among them. Many of the later actions possess an array of original adjustment mechanisms that require skill and experience to exploit to advantage. A seemingly slight change in settings can produce a panoply of unintentional or unexpected effects. Actions that are not fitted with obvious adjustment points should be approached with even greater care, since changes are less easily reversed. The subtle interactions and relationships among the moving parts cannot always be appreciated simply by peering at drawings, or even at the actions themselves. Experimentation can be enlightening, but damaging. Specialist advice should be sought whenever doubt arises about the causes of problems or possible solutions.

Up

General Guidelines

A well set up action, whether harpsichord, clavichord or piano, is, above all, reliable in its behaviour. A few guidelines apply to all well-regulated pianos. Any change in touch or tone, or failure to perform according to the general guidelines below, may indicate the need for attention. There may be differences between the performance capabilities of square and grand pianos, but the specific variations are beyond the scope of this section.

Pianos, of course, possess struck actions, but unlike clavichords, the striking part, the hammer, on the majority of actions, is designed to be free-flying when it meets the strings. In such actions, on a very slow depression of the key, the hammer should "let off" before it reaches the strings ­ that is, it should not reach the strings ­ otherwise it is said to be blocking. The let-off point should not vary appreciably on adjacent notes, though it may change gradually from bass to treble.

A hammer should not bounce against the strings (play more than once on one striking), even when the key is struck at the loud limit of its dynamic range, and even when there are no checks. On check-less actions, the dynamic range may be narrower than on instruments possessing checks. If a hammer bounces on a well-regulated checkless piano, then it is being struck too hard. By the same token, notes should be able capable of very rapid repetition at all intrinsic dynamic levels.

When pedals, knee-levers or hand stops are not in use, the hammer head should strike all the strings associated with its particular note centrally. Pedals and knee levers should operate quietly, and without undue resistance.

The whole action is either fixed or ultimately resting upon a piano's keyboard. Any upset in key level can significantly alter the setting up of sensitive early piano actions. Specialist advice should be sought whenever any aspect of a piano's regulation requires attention. Regulation on these, and all keyboard instruments, is an iterative process which may be very time-consuming. Changes in regulation can be expected as an instrument is played in, as temperature and humidity vary, and as a result of wear and tear engendered by use.

There are many types of early piano actions, but a large proportion can be described as resembling, in important respects, the three mentioned below.

The Cristofori piano action was invented by Bartolomeo Cristofori around the turn of the 17th century. It was so highly developed and ahead of its time that much of the subsequent history of the piano action is a re-invention or re-discovery of his work. Equally remarkably, he recognized virtually every significant structural requirement of an instrument that was to possess struck rather than plucked strings. The materials upon which the action relies most heavily are wood, leather and cloth, although it also includes wire springs, stems and pivots. Cristofori intended adjustments to be made by bending metal parts or varying the type and thickness of the cloths and leathers. However, no such adjustments should be undertaken on historic instruments without reference to specialist advice.

Viennese actions are often referred to as the Prellmechanik, and a growing body of information is now available on the history and constituents of the actions that would have been the most familiar to Haydn, Mozart and Beethoven. In essence, it is an action in which the hammers are pulled (or flicked) into motion, as distinct from the English and Cristofori actions where they are pushed. During the course of the 19th century, the Prellmechanik, which had existed side-by-side with the pushed type of action, was gradually supplanted by the latter. The materials used in the action include wood, leather, vellum or parchment, brass and steel, cloth and felt. Regulation procedures are highly instrument-specific. Routine maintenance should rest with someone expert in dealing with the idiosyncrasies of each instrument.

English actions contain the same materials as found in the Viennese instruments described above. Pianos by Broadwood, which were made and, fortunately, survive in relatively great number, reflect a certain standardization, since principles of mass production were employed in their manufacture. Nevertheless, during the long history of the firm, many changes in compass, action design and materials took place. These have still to be adequately chronicled. Any actions, in grand or square pianos, which largely or in part, retain original materials, should be carefully documented and conserved.

Up

Care of Ephemeral Parts

Strings

String breakage can be kept to a minimum by maintaining stable environmental conditions, and allowing the pitch to wander only within narrow limits (in the region of 3 Hz above or below its norm at reference pitch). Persistent breakage within an instrument, or on a particular note, suggests an incompatibility among the pitch at which the instrument rests, its scaling, and the string material chosen for the offending note(s). Such problems should be rectified by a specialist. Unavoidable breakage usually results from metal fatigue, and occurs at the string's stress points ­ where it leaves the tuning pin for the nut, at either end of the hitchpin loop winding, or where it passes over a pin. As with tuning, the techniques of string replacement are far more various in historical instruments than in modern pianos. When making and putting on a new string, care is taken to replicate as nearly as possible the size and style of the hitchpin loop and tuning pin coil, as well as the material and diameter of the old string. A list of string materials, diameters and overlengths may exist from the time of the instrument's last treatment, and surrounding strings can often be a guide to correct downbearing, and tuning pin height. Original strings, or fragments of them, should always be meticulously preserved. Historical reproduction music wire may or may not be metallurgically similar to historic wire, but more important is that its tensile strength lies within the same limits.

Up

Plectra

The materials most often used for harpsichord plectra are natural bird's quill and leather. The longevity of each is directly proportional to how much it is used. The active life of these natural materials can be prolonged significantly by simple maintenance procedures.

Both sides of quill plectra require light lubrication from time to time, so that the top surfaces pluck the strings with the correct amount of friction, and the bottom surfaces re-pass the strings without catching (hanging). The oil on one's own skin is excellent for these purposes, and always to hand, so to speak. It is advisable to check quills both before and after any use. The top surface may need lubrication if the order or spacing of the stagger is upset, if there is an extraordinary resistance to the touch compared with surrounding notes, or if a note sounds peculiarly loud or produces a small grinding sound when flexed against the string. Usually, all of these symptoms are present when light lubrication is required. The offending quill should not be pushed past its string until lubrication is applied. The temptation to lubricate regularly quill plectra to solve no specific problem should be resisted because lubricants attract dust. Too much lubrication also makes strings, and any other parts of the action which come into contact with the plectra, sticky.

Leather plectra can also require lubrication for the same reasons as quill, but even less frequently. Incautiously lubricating leather can spoil the sound, so it should be applied sparingly. Hard leather plectra, which have begun audibly to soften, can sometimes be revitalized by applying, with a fine brush, a minute drop of weak shellac solution to the underside, at the root. This is a task for an expert as the dilution of the shellac is critical and even the slightest excess can ruin the plectrum or the tongue.

The cutting, shaping and thinning of any plectrum material requires experienced hands. Further advice should always be sought when plectra require modification or replacement.

Up

Cloths and Felts

The information on textiles in Chapter 5 applies equally well to instruments in regular use. If an instrument has rare, original or old cloths, regular inspection and preventive cleaning should be scheduled if necessary. Sensitivity and a long-term perspective is required whenever any decision is made to clean. The process itself invariably removes some of the very material which is being preserved, together with unwanted dirt.

Up

Leathers and Parchments

These materials perform a large variety of functions in early keyboard instruments, as explained above. Like cloths, they are used not only in "passive" roles such as for silencing noise, but also in "active" ones such as damping, the hinging of moving parts, and plucking strings. Many leathers must withstand friction (like those used on check leathers), and repeated battering (hammer coverings), and also be able to resist persistent flexing (beak leathers and leather plectra). In many of the above functions and others, leathers define dimensions critical to the touch and repetition of the action, while the characteristics of the leathers used for plectra, hammer coverings, and dampers largely determine the nature of the tone and decay of notes. Many of the leathers of the type and with the attributes of those used on historic instruments are practically unobtainable today.

The cautions and provisos regarding the cleaning and general treatment of leather and skin products apply equally well to functioning instruments.

Chapter Head

String Instruments

It is commonly argued that string musical instruments must be kept in playing condition, and that they deteriorate while on display or in storage. Some violin collections, such as at Cremona, are governed by a policy of daily playing in order to keep them in good condition. There is no supporting evidence for the benefits of this practice, but it is reliably known that the musical qualities of long unused instruments are in no way compromised by their silence.

Should a string instrument have already been brought into stable, playing condition, and should there be a policy that it continues its playing role, it must be regularly maintained, tuned, and adjusted. Unlike those on display or in storage, playable instruments should be kept under tension at almost the appropriate playing pitch. Maintaining historic instruments in this state is the province of those skilled both in modern techniques of practical music and in museum standards.

Up

Preparation

The adhesive traditionally used for wooden instruments is a collagen-based glue made from hides, skins and bones. It is normally very strong, but is susceptible to degradation by exposure to moisture and continuous stress. All glue joints must be rigorously examined for signs of weakness and failure. This is a highly specialized undertaking. Open cracks or joints should indicate the need for lowering string tension and consultation with a conservator.

The strings found on historic instruments are rarely original, except in the case of recently made instruments, and may often be a different gauge than those originally fitted. Nevertheless, before any restringing is done, the nature and history of the existing strings must be ascertained. Strings that have been removed should never be discarded. The material of any strings newly put on an instrument should carefully be considered. On instruments of the violin, the viola da gamba and lute families, it was standard to use gut even for the uppermost strings, while metal covering of the lowest gut strings is only known since the second half of the 16th century. Strings made totally of metal did not appear before the 1920s. In addition to the material of strings, their production techniques have also undergone considerable development, in particular since the 18th century. Before that time pictorial and written sources show that the lower strings were often twisted from several strands. Such considerations are equally valid for the metal stringing on citterns, mandolins and other types of instruments. Therefore, it should not be assumed that the use of strings made of traditional materials will be safe for the instrument, or will lead automatically to an authentic sound. It is dangerous, and pointless to string an instrument if the original stringing material and tuning are unknown. Also, the pitch at which the instrument was normally tuned should be respected; a tension of even a half tone higher than the original value can have marked deleterious effects. No historic string instrument should be brought up to playing tension in one session; it is necessary to tighten the strings to some previously established low tension and to assess the effect over a period of at least a day. Extra time should be allowed for more sensitive instruments, like those of the violin and viol family. Tensioning continues incrementally until stability at the correct pitch is attained. The type of instrument, its condition, the original pitch and stringing, and many other interrelated factors make specifications impossible to establish.

All tuning devices should be checked for smooth operation and efficiency. The directions provided in Chapter 5 apply equally to playing instruments.

Up

Care During and After Use

Highly finished wooden instruments are very susceptible to damage while in use, and every precaution should be taken to ensure that they are handled carefully and that players are fully aware of their fragility and individual foibles. Surfaces should be protected from abrasive contact with the player's clothing, particularly buttons, zip fasteners and buckles, by using soft cloths or other padding.

Players should not be permitted to change the set up of instruments unless under direct supervision or consultation with qualified personnel employed by the museum. The instrument should not be used in a manner inconsistent with either its historic state or with the period it represents. Players should not be permitted to use anachronistic and potentially harmful items (such as chin or shoulder rests for violins) that might cause damage.

When bringing a bowed string instrument to its initial pitch, and when subsequently tuning it, the bridge might alter its angle in relation to the table (when seen from the side). Readjusting this angle should be left to an experienced conservator. Similarly, although the player may have suggestions concerning the adjustment of the soundpost position, this is also a task for a conservator well acquainted with bowed instruments. While the use of rosin is necessary when playing instruments of the violin or viola da gamba families, it is easily deposited on the varnish near the instrument's bridge. An agreement should be reached between the conservator and the player concerning who is responsible for removing rosin. Historical bows that have retained hair from any earlier period should not be used. Such bows may preserve unique data regarding the nature and mounting of the hair. Instead, the player should provide a bow appropriate to the period and style of the instrument.

With plucked string instruments, the relation between the strength of the instrument's construction and the total tension of strings is particularly subtle. For this reason, early lutes are rarely able to support the stresses of playing. If this is done, it should only take place in the presence of a conservator who has specific experience. This also applies to guitars, mandolins and other plucked instruments. Harps are especially difficult because their frame construction is extremely liable to distortion.

As with all other instruments, thorough and accurate records should be kept every time an instrument is used.

Chapter Head

Woodwinds

Up

The Materials of Fabrication

Up

Conditioning Before Use

Also, historical instruments that have survived brief periods of usage often enough to be regarded as "safe" can quite surprisingly crack under what would not be regarded as unusual circumstances. Thus, it may also be reasonable to conclude that the final irretrievable strain that causes a crack to form in an old instrument is part of a cumulative process. It would follow from this that there is no way to know when a woodwind instrument is likely to crack, and there is, therefore, no adequate way to guard against that eventuality. The minimum risk condition is to allow less valuable instruments that are candidates for playing to go into equilibrium with upper range of relative humidity and temperature values. When the moment of ultimate risk is at hand, the instrument can be further raised towards body temperature (tucking it between the body and the upper arm is a tried and true woodwind musician's device) before playing.

Up

Care After Use

Cleaning after use is essential. In general, the processes described in Chapter 5 are equally appropriate for working instruments. As with all other instruments, woodwinds should be documented fully before, during and after every use, and the written record used as a guide to assessment of their continuing playable condition.

Chapter Head

Brass Instruments

Up

The Materials of Fabrication

Up

The Joins Between Components

Up

The Surface

Up

The Interior

Up

Movable Parts

No instrument should be disassembled without using the correct tools. It is very rare that modern, standardized tools can be used on early instruments. It is often necessary to make tools for specific tasks, or to adapt commercially made ones. Disassembly and reassembly require specific skills and techniques and a dedicated workspace.

All excess deposits of old lubricants and accumulated dirt should be removed using a mild solvent, such as turpentine substitute applied with a small cotton swab. Parts such as tuning slides, which remain static while playing, should be moved periodically while in storage or on display to avoid seizing. Seizing results when metals remain in close contact with each other for long periods. Parts that are designed to be tight fitting, like tuning slides, are most prone to this. Similar metals, such as brass on brass, are more likely to seize than dissimilar metals, like nickel-plated slide stockings acting on brass tubing, for example. Effective lubrication of such parts is a key to their remaining movable.

The purpose of a lubricant is to provide a thin film of fluid that acts as a barrier to prevent direct metal-to-metal contact between moving surfaces. No valve, keywork, piston, or other moving part of a brass instrument should be operated dry. The type of lubricant used on valves and slides depends critically upon the closeness of tolerance of the components. Working surfaces of older instruments may need very different kinds of lubricants from those used on new ones. Also, lubricants designed and marketed for frequent and continuous use may be inappropriate for less arduous applications. The optimum viscosity to be used for valves in good condition is between 2.5 and 4.0 cSt, but badly worn valves can tolerate or even benefit from somewhat higher viscosity oils. However, if a lubricant is required for badly worn valves, the question of whether the instrument should be played at all arises. As a general rule, a lubricant should be selected within the viscosity range above, and should contain no added solvents. Silicone products have been considered taboo in museum conservation for a variety of reasons, most of which concern their use on wooden surfaces. There is no reason to prohibit their use on metal slides; in fact they have characteristics of long-term stability and easy removal that make them superior to mineral- or animal-based oils.

Movable but non-moving parts, such as tuning slides, should be coated with a lubricant of much higher viscosity, because the low viscosity lubricants recommended for moving parts can flow away from surfaces over long periods of time. A silicone grease very sparingly applied can be recommended. Routine inspection at yearly intervals will ensure that the lubricant is still doing its job. If components are extremely tight, and there is the risk of causing damage by moving them, wherever possible they should be left disassembled.

Up

Care After Use

Chapter Head

Percussion

Chapter Head

Selection and Monitoring of Players

Sensitivity towards historic instruments is an essential attribute of a player's training. Those who wish to use the instruments in a museum collection should be required to establish their credentials and qualifications. An expressed interest in early instruments is a starting point, but this must be underscored by experience and knowledge specific to the instrument under study. An exception may be allowed for young musicians because, having less training, they are more receptive to new ideas and can benefit over a long period from such exposure.

Monitoring the use of the instruments is an ongoing task, and it becomes an opportunity to establish a three-way dialogue among the player, the instrument and the person who provides the care. The latter must be well trained in all aspects of the instruments' function and use, and it must be understood by players from the outset that authority resides in this person. Written memoranda of understanding should always be made and copies added to an instrument's object file.

The international musical instrument committee (CIMCIM) of ICOM has tackled this problem in Recommendations for Regulating the Access to Musical Instruments in Public Collections. This document was written by a specially convened group of musical instrument conservators and curators, and published by CIMCIM in 1985. It establishes conditions and demonstrates current approaches for allowing museum visitors (instrument makers, historians, musicologists, players, etc.) to physically handle museum musical instruments. The document describes the conditions of access, general protection from damage, measuring tools and techniques, and the conditions for playing, while also touching upon the question of copyright.

Chapter Head

Relative Costs

If maintenance and regulation is to be an in-house affair, it is necessary to retain a person on staff with the full capabilities and qualifications for the task. Very often old instruments sound and feel very differently from later counterparts, and since it is often necessary to demonstrate techniques to would-be players, an ability to play the instruments competently is a huge asset. In addition, the incumbent must be fully familiar with the construction, conservation and maintenance of the specific types of instruments in the collection. The position will need to be paid at a reasonable rate for such specialized technical work, and must be supported with a well-equipped workshop and monitoring instruments.

If maintenance, regulation, and general overview of the collection is to be contracted out, the frequency and length of visits, the extent of work performed and, of course, fees paid, should all be established in advance. It is also necessary to factor in the costs of preparatory work on the instrument, monitoring of the playing session, and post-playing maintenance and inspection. Also, potential players must be introduced to the instruments and auditioned to ensure that they are capable of working safely with historic material. Such introductions can be costly and time-consuming, but for the well-being of the instruments they are indispensable.

Chapter Head

Retiring Instruments from Playing Status

Retiring from playing status requires acute discrimination of exactly when the instrument has reached the point where too much intervention would be required to keep it working. Nevertheless, many functioning instruments contain little of irreplaceable historic value. It may best serve the collection if they continue to be maintained in playing condition. Because retiring from playing status is a highly complex and multi-levelled decision-making process, every effort must be made to preserve the results of the deliberations from the arbitrary. All points of view should be considered at the time the decision is made, and a permanent record of the deliberations should be prepared for inclusion in the instrument's object file.

Chapter Head

Conclusion

Karp has stated that:

It is becoming obvious that restored older instruments do not necessarily behave or sound as they did when they were new. Since composers can reasonably have expected to hear their works performed on relatively new instruments ­ certainly not on 200 year old restored museum pieces ­ the chronologically original instrument is not necessarily the musically authentic one.

There is an apparent reluctance among custodians of historic instruments to accept the fact that the sound quality that an instrument at present possesses may be widely different from its original quality, and that no amount of introspection and scholarship can capture its true original state. However, it is generally recognized that playing instruments that are in sound enough condition to support it is, and will continue to be, an attractive interpretive function regardless of the veracity of the results.

Chapter Head

Return to Contents

Chapter Head

General Principles

Documentation is so important that even in small museums and private collections procedures should be codified and made independent of individual people's memories. As in other areas covered by this book, the best documentation practices of private collectors can be up to the standards of professional museum management.

Chapter Head

Registration

Registration methods are described in a number of publications cited in the bibliography and will be only summarized here. There is nothing special about registration of musical instruments, except to say that it is sensible for items that function independently to be allocated distinct acquisition numbers. For example, bows for stringed instruments, mouthpieces for brass instruments, and beaters for percussion instruments can be given their own numbers, even when associated with instruments. The nature of associations between items should always be recorded. Cases, ephemeral parts, and accessories rarely need separate numbers if entering the collection associated with instruments, but can be dealt with in documentation along with those instruments. Consistency and accuracy are more important in registration than using technical terms (which may only be learned by the registrar at a later date).

A permanent accession register of all items in the collection should be maintained in order of acquisition number. This is the master list for documentation purposes. For each item, an entry should be made containing date of acquisition, the unique acquisition number, a brief description including the object's own serial number (if any), whether donated, purchased or loaned and name of donor, vendor, or lender. Information should be added to the register, or a separate register made, to record the withdrawal of loaned items and ownership changes of loaned items.

Objects entering the collection should be labelled with their acquisition numbers. Tie-on tags or similar removable methods are preferred for museum material, but if numbers must be applied to the items themselves, they should not be marked by irreversible means. Marking of objects should be durable, if not indelible (Barclay, Care, p.11).

Chapter Head

Object Files

• valuations for insurance purposes;

• location lists;

• display label texts;

• catalogue descriptions of the object and museums research reports;

• photographic and radiographic records; and

• records of loans from the collection.

Care plans containing the following information are another vital component of documentation:

• records of conservation treatment and repairs carried out by the museum;

• condition reports;

• audit records;

• playing logs; and

• condition monitoring reports.

The last four categories are dealt with further in more detail.

The files outlined in the two groups above will not be organized in order of accession number, but the organization of the files should enable all the information held about a particular object to be readily identified. For non-computerized documentation systems, however, indexes may be necessary for information retrieval. These indexes have conventionally been card files ­ either subsidiary indexes or duplicates of master cards that are kept in different sequences.

Other documents connected with objects may include drawings, makers' trade brochures and instructions, patent specifications, written, audio or video records of the instrument in use, interviews with makers, players and owners, etc.

Chapter Head

Security

Chapter Head

Cataloguing

Wherever appropriate, cataloguing standards should be adopted. The structure of the documentation may be dictated by a system (such as Spectrum in the United Kingdom), or a regional co-operative scheme. For specialized collections, published catalogues can, in many cases, be used as models (see, for example, Borders, Gunji, Heyde, Koster and Myers). A cataloguing structure suitable for a general collection of musical instruments, and equally suitable for electronic and manual formats, can be found in CIMCIM Newsletter XIV, 1989.

Standards are necessary to ensure consistency in presentation when a number of consultant authors are involved in preparing catalogue copy. The main advantage of computerized systems is that they facilitate revising the data when justified by additions to the collection or by new information about items already held. Another advantage is that collected information can be manipulated and filtered for different purposes without complete re-writing. However, if a computer is not immediately available the documentation processes should not be deferred: data can be keyed or scanned into electronic form at a later date.

A catalogue of museum instruments should relate the knowledge of the craftsmanship of instrument making to the knowledge of historical practice in music making. The description should not only include the physical description of objects, but also an indication of the resources provided by instruments to performers during their working timespan. It is the synthesis of these elements that makes the documentation requirements of musical instruments different from those of non-functioning museum objects. Typical readers of catalogues are players of instruments and instrument makers. The former are interested in the characteristics of historical models and require information that places the instrument in its musical, cultural, and social context, whereas makers require information of a different kind.

In a relatively well-defined area of museum activity such as musical instruments, it is a realistic goal to harmonize methods in order to improve communication. The harmonization of cataloguing methods has for a long time been a feature of professional librarianship, which has developed recognized international standards for both describing and ordering of catalogue information. These standards structure the recorded information about published books and other materials into precisely defined fields, with procedures for recording the information in cases capable of differing interpretation. Standardization ensures that the same book, catalogued by librarians working in different places, will be given identical, or at least compatible, descriptions. For museum documentation, standardization is advantageous for:

• procedures

• record structure and ordering

• terminology

A standardized approach is, of course, made more possible and more necessary by adopting computers and electronic data communication methods, but it is equally feasible for a paper-based system. The systematic, routine recording of information and the production of a published catalogue can be separate processes. Even if a published catalogue is not a high priority, data need to be recorded in a clear, precise way so that eventual publication, on paper or electronically, will be well prepared for.

It is impossible to prescribe cataloguing rules that will apply to every single instrument; cataloguers must approach exceptional cases by following the underlying principles of the cataloguing standards. Standard rules should be hospitable to changes in emphasis within categories of information over a period of time, and to the introduction of new categories if and when any aspects of the description of instruments become more interesting to scholars.

Because they are designed to play music, musical instruments do not always conveniently fit into the schemes of classification by which musicologists and curators arrange their information. Instruments can be classified according to their place, time and culture of origin, acoustical principles, their morphology, their materials of construction, their function (either at time of manufacture or at the time of their most recent regular use), etc. Any classification bringing together instruments that share one characteristic will separate instruments that share other characteristics. The most widely used classification is that of Hornbostel and Sachs, which uses acoustic criteria for its primary division and morphology for secondary divisions. This is the most appropriate ordering for the documentation of many collections. A useful supplementary general rule for ordering a catalogue is that similar instruments within a class (e.g., transverse flutes of whatever keywork) should be listed in order of descending nominal pitch.

An instrument should be described in the functional state in which it left the workshop of its last repairer or, if in original condition, its maker. Any faults, damages or losses that have occurred since then can be listed with a description of the object's condition. Although it is interesting to speculate about former states of a modified instrument, these opinions should not be confused with the description of the instrument's condition, but should be given with any other description of repairs to the object. For example, if a bassoon was built as an eight-key instrument, had a ninth key added later, and has since lost a couple of keys, it should be catalogued as a nine-key bassoon; the loss of two keys is described along with other aspects of its condition, and the addition of one key described along with other repairs.

A concise, useful system for describing keywork of wind instruments is that developed by John Dick. This system does not attempt to analyze the mechanism from the engineering point of view, or to describe the keywork fully, but rather to indicate the facilities offered to the player. Where measurements are valuable in indicating playing characteristics, such as drum diameters, trombone bores, guitar string lengths and violin body sizes, these are worth giving.

All measurements should be given in international (SI) units. Consequently, physical dimensions of musical instruments should be given in millimetres or metres only. Cataloguers should state the degree of accuracy associated with their measurements, whether length, pitch, or any quantity expressed numerically. For example, if the diameter of a woodwind finger-hole is stated to be 2.25mm, it should be clear from the text whether it is within the range +/-0.01mm or +/-0.25mm. Where appropriate, the equilibrium temperature and relative humidity at the time of measuring should be stated. If measurements in units in use at the time and place of the instrument's manufacture are thought to be significant, these may follow the data in SI units.

Although some museums still use the Helmholtz pitch nomenclature, it is preferable to adopt the internationally-used American National Standard nomenclature. The note letter is followed by a number (where possible written as a subscript) denoting the octave. C4 to B4 is the octave from "middle" C to B in the centre of the treble clef; higher octaves are denoted by higher numbers, lower by lower. Thus the A that is commonly 440 Hz is A4. The cataloguer should make clear wherever necessary whether the actual or the transposed ("written") pitch is meant. For regional instruments where international pitch nomenclature is inappropriate, pitches can be given in terms of the ellis (El). The octave number (where possible written as a subscript) is followed by the pitch in cents above the nearest C below the pitch being described. Thus the pitch of "middle" C is 4/0 El; 440 Hz is 4/900 El. Where the information given is only the opinion or speculation of the cataloguer, it should be clearly indicated, for example by a question mark in parentheses (?).

There is no universally accepted thesaurus of terms used in the study of musical instruments, so in the case of variants, the choice of name or of spelling should follow the usage of The New Grove Dictionary of Musical Instruments.

The common usage of auctioneers is also applicable in museum catalogue descriptions:

ascribed to

A traditional attribution with which the cataloguer does not necessarily agree.

attributed to

The instrument is believed to be by the named maker in the opinion of the author(s) or the authority(ies) whose literature or certificates are cited.

by

The instrument is in the cataloguer's opinion the work of the named maker. This category also includes instruments made specially for a dealer and originally sold under the maker's name, and where the actual maker is unknown or unidentifiable.

labelled, stamped, inscribed, etc.

The instrument is not, in the cataloguer's opinion, by the maker indicated but merely bears his or her name. In some cases the instrument may be a later copy or be modelled after the maker indicated.

school of, ...school

The instrument is, in the opinion of the cataloguer, made by a follower of the maker indicated, or is in the style of instruments associated with the area indicated.

workshop of

In the cataloguer's opinion, the instrument is executed in the basic style of the maker and possibly under direct supervision.

The text of any inscription should be prefaced by the method of inscription (such as branded, inscribed, carved, engraved, labelled, stamped, written, etc.). A useful convention is to transcribe the text of the inscription in double-quotes. Line breaks in the original are indicated by a oblique (/). Any trademarks, logos or devices are described outwith the double-quotes, but in sequence of line breaks. If the whole inscription is not given, parts omitted should be indicated by "...". If parts are illegible, they should be indicated by a convention such as "(­ - ­)". The location(s) of any inscription on the instrument should be indicated. In the case of transliteration from scripts other than the language of the catalogue, the standard orthography used should be cited.

A reference should be made to any other item in the collection directly associated with the item being catalogued. This is standard practice for brass instrument mouthpieces, bows of stringed instruments, beaters of percussion instruments, and instruments built, sold or used in pairs or groups. A cataloguer should indicate whether the association is by design, manufacture, sale or usage, and whether long-standing or recent. In many cases, the curator does not know the nature of the association, but may have good reasons for making an informed guess; in these cases, the guess can also be valuable and should be recorded with some indication of its degree of uncertainty such "possibly", "probably", etc.

When dealing with playing adjuncts, the editorial policy is not necessarily obvious. Some adjuncts are used by players for more than one instrument, and catalogue users may find it helpful to have them placed together in the catalogue so that they can make the kind of comparisons that players make when selecting their equipment. This consideration has to be taken into account when deciding a cataloguing policy for the following:

• bows for stringed instruments (not usually associated by makers);

• beaters for percussion and plectra for stringed instruments (the degree of association varies);

• mouthpieces for brass instruments (often associated by makers of instruments, but commonly provided by different makers);

• crooks for bassoons, horns, etc., and tuning bits (usually provided by instrument makers but commonly dissociated by owners);

• mutes (rarely supplied with or specifically designed for an instrument); and

• strings, drumskins, reeds (ephemeral: any long-standing association is a valuable rarity).

For these, and no doubt for other categories, the editor of a catalogue must consider the size and scope of the collection and the needs of catalogue users. It is suggested that of the above categories, brass instrument mouthpieces, percussion instrument beaters, and bows for stringed instruments are frequently used for playing various brass, percussion and stringed instruments and should be catalogued separately even when associated by manufacture or long usage with particular instruments. Other playing adjuncts can be described in the catalogue as accessories for the instrument they have been associated with most recently, if any. In all cases, the cataloguer should state whether the accessories are known to be those originally supplied by the maker for a particular instrument, whether they are probably original, or whether they are only possibly original.

Published literature and recordings referring to the specific item being catalogued should be cited. Known usage in a particular orchestra, band, etc. should be stated, together with players' names and dates. The names of former owners, collectors and players (in reverse chronological order with dates) should be given. Finally, the names of the cataloguer(s) and date(s) of working should be recorded as part of the catalogue documentation.

Chapter Head

Documentation and Monitoring of Instruments

The sub-sets of documentation within the care plan relating to the use of instruments involve four separate kinds of records: condition reports, sounding and playing logs, monitoring logs, and audit records. All such documents must be kept in the object file unique to each item of the collection. Although details vary with the type of instrument and the circumstances, the following guidelines may help determine the kind of information that should appear in these records.

Up

Condition Report

Up

Sounding Log

• date of audit;

• auditor(s);

• details on preparation of instrument for audit;

• information on environment;

• duration of audit;

• results obtained; and

• details on treatment of instrument after audit.

Up

Playing Log

• date of playing;

• player(s);

• condition report before use concentrating on working parts;

• details on preparation of instrument for playing;

• duration of playing;

• information on transportation and environment;

• condition report after use, concentrating on working parts;

• details on treatment of instrument after playing;

• recommendations for future use; and

• information on recording, broadcast, etc.

Up

Monitoring

Up

Audit Records

Chapter Head

Automated Resources

The process of electronic documentation is divided into the separate stages of data capture, processing, and dissemination. The capture of digital data consists of such activities as entering the description of an object into a database, scanning a photograph, or making a digital recording of the sound of a musical instrument. The processing of such data might include a database search, transforming a digitized photograph into a line drawing, or the spectral analysis of a sound recording. Dissemination could include providing access to a data base system via the World Wide Web, posting a digital image via electronic mail, or selling a compact disk in a museum gift shop.

Up

Databases

It subsequently became apparent that there were more efficient and reliable models for structuring the data in automated systems. The development of data modelling methodologies soon became a specialized discipline; one which still retains its importance. At the outset, an overriding concern was the high cost of magnetic storage media. It was common to use terse abbreviations and codes whenever possible in order to minimize the need for space on disks and tape. Redundancy in the stored data was also shunned.

The most often mentioned stages in the development of methodologies are the hierarchical, network, and relational models. The last of these is the only one based on a coherent mathematical theory, and it serves as the basis for a large proportion of present-day database management systems. By definition, the relational model only permits data to be expressed in tables, with a database being nothing more than a set of interrelated tables. (The term "table" is a synonym for "relation." A relational system is so called for that reason alone. There is a popular misconception that a system is termed relational because it reveals relationships.)

For example, one table in a database might state and define the names of all indigenous western Canadian musical instruments. Another would provide biographical data about the makers of these instruments. A third might describe the materials used in the instruments' manufacture. A fourth could describe a specific collection of instruments, including references to relevant data in the other tables. The important feature here is that data relevant to (for example) the Haida rattles would be listed only once in the entire database. The record about each of the many instruments made by the Haida in a given collection would contain a brief reference to the nformation about them in the biographic table. When structuring a relational system, it is important to note that such things as a maker's name are attributes of a musical instrument. A maker's place of birth is not an attribute of an instrument; it is an attribute of the maker.

In addition to reducing redundancy, the relational approach also enables validation control. If, for example, a collection contained 357 Haida rattles and Haida was listed in the database record for each of them, there would be a large risk of some spelling mistakes. This could, in turn, result in a database search for a spurious object because it was attributed to the Haida by virtue of a spelling error. If at some point research were to reveal that Haida actually should be spelled Haidda, making the appropriate change to a single record in the biographical table would automatically correct the entire database. If there were multiple occurrences of the name throughout the database each would first have to be located, including all inadvertent spelling variations, and then corrected, risking additional misspellings.

Were this section being written two years ago, the only modelling methodology likely to be discussed would be the relational approach. In the interim, however, a significant portion of the museum community has begun to use the Internet as a presentation platform for documentation systems. The software resources normally encountered on the Internet provide little of the sophistication that database management specialists have now come to expect. Not surprisingly, the developers of networking tools have traditionally placed greatest emphasis on communication facilities.

From the database specialist's point of view, the World Wide Web is a major step backwards to a time when the "network model" presented all sorts of difficulties. As will be explained in greater detail below, a definitive characteristic of the Web is that it contains countless documents that provide links to each other. If whatever a link points to should subsequently be changed, or even deleted, there is no mechanism to ensure that a corresponding change is made to the link. Nor can all copies of a document that are maintained under the same name at numerous physical locations, be controlled or remain identical. To put it another way, if a certain museum Web site were discontinued, the many sites which link to it might or might not be informed of this, and thus their data would be faulty.

Some relief is provided by relational systems that support access via the Web. However, such systems are not immune from a higher level version of the problems just listed about the referential integrity of individual documents. It can never be certain that what initially was a link to a relational database will remain so. There are, nonetheless, significant advantages to be gained by freeing individual database systems from these problems.

Up

Research Instrumentation

Creating a digital record also involves the loss of data, this time by definition. In the analog record (a drawing), a single straight line is often used to connect the end points of a surface that is assumed initially to have been at least conceptually straight. A digital record can define that surface in terms of a far larger number of points than just two. So-called 3-D scanners are capable of making contact-free measurements, translating an object into a large number of digital coordinates. These data may then be used to generate everything from technical drawings to the direct control of an automated replication process. Although computerized scanners of this type are not yet available on a consumer level, they are encountered in larger museum installations. There are, however, a variety of low cost digitizing devices that may easily be incorporated into a modestly funded documentation facility. These range from digital callipers to a variety of smaller 3-D scanners, all of which can be attached directly to standard microcomputers.

Facilities for making digital sound recordings are now commonly included in "multimedia" packages for consumer-level personal computers. A "sound card" together with a reasonably inexpensive microphone can easily provide a digital record of an instrument's acoustical output. The software normally provided with such equipment can allow the processing and analysis of these recordings in a way that was formerly regarded as utopian. Similar means are also available for digitizing videotaped performance events. At present, however, the storage requirements for all but the briefest digitized video sequences can be daunting.

Chapter Head

The Internet

The essence of the Web is what are called "hypertext" documents. These can be understood in terms of the way a conventionally printed book is read. Someone can decide to turn from page 52, on which a reference to a potentially interesting footnote is provided, to page 312 to see the extensive footnote text. In that footnote, reference can be made to another book, which the reader can decide to consult prior to continuing with the first book. In the second book, reference might be made to a videotape which is then viewed before returning to the printed page. A soundbite on the videotape could trigger the desire to listen to a CD containing a full performance of whatever the snippet triggered an interest in. A hypertext document may be visualized in the same way, as a digitized book that provides a convenient means for jumping from place to place within it, and which contains multiple media ­ sound, video, graphics, etc. If the computer on which this book is stored can establish networked connections to other computers that contain hypertext documents, we can add convenient links to them.

Controversy has long raged about the proper focus for the documentation of a musical instrument. Should the primary concern be placed on its morphological description, its acoustical properties, its use in a musical context, its function in a cultural/social complex, etc.? With the advent of digital multimedia applications it is possible to create a single documentary record of any given instrument and include as many of these facets as appropriate. The use of hypermedia networking facilities allows such documents to be linked into a global aggregate.

Although museums obviously will remain the hosts for their collections, access to the documentary records of these collections no longer needs to be seen in terms of physical access to the museums. As digital imaging techniques become more powerful, the need to visit a museum to see an object may become less acute. Similarly, useful "here's what it sounds like" data can be provided, thus reducing the need to play instruments.

Chapter Head

Readings in Electronic Documentation

Up

Database Modelling and Design

Up

Electronic Instrumentation

Everything relating to the Internet ­ from recreational use, to descriptions of the most intricate services ­ is described in a multitude of publications, the largest part of which is available without cost on the Internet itself. Because the first encounter with the Internet can be confusing, the most important attribute of any book on the subject is its personal appeal to the reader. The most highly regarded single work is The Whole Internet User's Guide and Catalog by Ed Krol. There are several network-based points of entry into the vast wealth of material that the Internet houses. The most useful single source of information relevant to musical instrument conservation is the website of the International Musical Instrument Committee of ICOM (CIMCIM) which contains useful information on a wide range of musical instrument museum activities:

http://www.cimcim.icom.museum

Chapter Head

Return to Contents

Arnold-Forster, K., and La Rue, H. Museums of Music. London: Museums & Galleries Commission, 1993.

Art History Information Program. Art and Architecture Thesaurus. Oxford: Oxford University Press, 1994.

Conservation Concerns: A Guide for Collectors and Curators. Washington and London: Smithsonian Institution Press, 1992.

Bacon, L. "The Deterioration of Four Giorgi Flutes Made of Ebonite and a Possible Method for Their Conservation." V. Todd, ed. Preprints for the UKIC 30th Anniversary Conference. London: UKIC, 1988, pp.96-100.

Barclay, R. "Ethics in the Conservation and Restoration of Early Brass Instruments." Historic Brass Society Journal, 1 (1989), pp.75-81.

Barclay, R.J. [sic] "The conservator: versatilty and flexibility." Museum International, 4 (1993), pp.35-40.

Barclay, R.L. "Instrument Mounts." Anatomy of an Exhibition: The Look of Music, Ottawa: International Institute for Conservation - Canadian Group, 1983, pp.101-109.

Barclay, R.L. The Art of the Trumpet-maker. Oxford: Oxford University Press, 1992.

Barclay, R.L., The Care of Musical Instruments in Canadian Collections. Technical Bulletin 4 (rev. ed.). Ottawa: Canadian Conservation Institute, 1982.

Barclay, R.L. "The Conservation of Musical Instruments." Museum International, 1 (1996), pp.9-14.

Barclay, R.L. "The Reconstruction of a Natural Trumpet by Sterner of Nürnberg." Galpin Society Journal, XLII (1989), pp.98-104.

Bordass B. Museum Collections in Industrial Buildings: A Selection and Adaptation Guide. London: Museums & Galleries Commission, 1996.

Borders, J.M. European and American wind and percussion instruments, catalogue of the Stearns Collection of Musical Instruments, University of Michigan. Ann Arbor: University of Michigan Press, 1988.

Canadian Conservation Institute, CCI Notes, Ottawa: Canadian Conservation Institute, 1982-.

Cassar, M., ed. Museums Environment Energy. London: Her Majesty's Stationery Office, 1994.

Cassar, M. Environmental Management: Guidelines for Museums and Galleries. London: Routledge/Museums & Galleries Commission, 1995.

Chartered Institution of Building Services Engineers. Lighting Guide: Lighting for Museums and Art Galleries, Guide LG/8. London: Chartered Institution of Building Services Engineers, 1994.

Codd, E.F. The Relational Model for Database Management, Version 2. Reading, MA: Addison-Wesley, 1990.

Comité International des Musées et Collections d'Instruments de Musique. Recommendations for the Conservation of Musical Instruments: an annotated bibliography. CIMCIM Publication No. 1. Paris: International Council of Museums, 1993.

Comité international des musées et collections d'instruments de musique. Recommendations for Regulating the Access to Musical Instruments in Public Collections. Paris: International Council of Museums, 1985.

Conservation Committee of the International Council of Museums. "The Conservator-Restorer: A Definition of the Profession." Bulletin of the International Council of Museums, 39, 1 (1986), pp.5-6.

Dawson, J., and Strang, T. Controlling Museum Fungal Problems. Technical Bulletin 12. Ottawa: Canadian Conservation Institute, 1991.

Dawson, J. and Strang, T. Controlling Vertebrate Pests in Museums. Technical Bulletin 13. Ottawa: Canadian Conservation Institute, 1991.

Dawson, J., and Strang, T. Solving Museum Insect Problems: Chemical Control. Technical Bulletin 15. Ottawa: Canadian Conservation Institute, 1992.

Dick, J.B. "Catalogue descriptions of instrument keywork," Galpin Society Journal, XLI (1988), pp.32-35.

Dudley, D.H., et al. Museum Registration Methods, 3rd ed. Washington, D.C.: American Association of Museums, 1979.

Elmasri, R., and Navathe, S.B. Fundamentals of Database Systems. Redwood City, CA.: Benjamin-Cummings, 1989.

Eliason, R.E., and Hellwig, F., eds. "Musical Instrument Exhibitions in Scandinavia." CIMCIM Newsletter special issue (1986).

Freeman, R. A Guide to Collections Records Management. Ottawa: Ministry of Citizenship and Culture, 1984.

Gilles, T., and Putt, N. The ABCs of Collections Care. Winnipeg: Manitoba Heritage Conservation Service, 1991.

Grant, A. SPECTRUM Essentials. Cambridge: Museum Documentation Association, 1994.

Gunji, S., et al., eds. The Collection of Musical Instruments. Tokyo: Kunitachi College of Music Research Institute, 1986.

Hellwig, F. "Packing and Air Transportation." Anatomy of an Exhibition: The Look of Music. Ottawa: International Institute for Conservation - Canadian Group, 1983, pp.63-73.

Heyde, H. Historische Musikinstrumente im Bachhaus Eisenach. Eisenach: Bachhaus, 1976.

Holm, S. Cataloguing made easy: how to use the MDA simple catalogue card. Cambridge: Museum Documentation Association, 1993.

Holm, S. Facts and artefacts: how to document a museum collection. Cambridge: Museum Documentation Association, 1991.

Hoadley, R.B. Understanding Wood: A Craftsman's Guide to Wood Technology. Newtown, CT: The Taunton Press, 1980.

Hoadley, R.B. Identifying Wood: Accurate Results with Simple Tools. Newtown, CT: The Taunton Press, 1990.

International Institute for Conservation - Canadian Group. Code of Ethics and Standards of Practice for Those Involved in the Conservation of Cultural Property in Canada, 2nd ed. Ottawa: International Institute for Conservation - Canadian Group, 1989.

Johnson, E.V., and Horgan, J.C. Museum Collection Storage. Paris: International Council of Museums, 1979.

Karp, C. "Restoration, Conservation, Repair and Maintenance." Early Music, 7, 1 (1979), pp.79-84.

Karp, C. "Storage Climates for Musical Instruments." Early Music, 10, 4 (1982), pp.469-476.

Karp, C. "Studies in the Documentation and Technology of Musical Instruments." Doctoral dissertation. Uppsala: University of Uppsala, 1984.

Karp, C., ed. The Conservation and Technology of Musical Instruments. Art and Archaeology Technical Abstracts. London and Los Angeles: International Institute for Conservation and Getty Conservation Institute, 28, 3, 1992.

Karp, C. "The Conservation of Musical Instruments." Per una carta Europea del restauro. E.F. Barassi and M. Laini, eds. Florence: Società Italiana di Musicologia, Leo S. Olschki, 1987, pp.283-289.

Kartomi, M.J. On Concepts and Classifications of Musical Instruments. London: Chicago University Press, 1990.

Koster, J. Keyboard Musical Instruments in the Museum of Fine Arts, Boston. Boston: The Museum of Fine Arts, 1994.

Krol, E. The Whole Internet User's Guide and Catalog, 2nd ed. Cambridge, MA: O'Reilly, 1994.

Kühn, H. "The Restoration of Historic Technical Artefacts, Scientific Instruments and Tools." The International Journal of Museum Management and Curatorship, 8 (1989), pp.389-405.

Light, R.B., et al. Museum documentation systems: developments and applications. London: Butterworths, 1986.

Lord B., and Dexter Lord G. The Manual of Museum Planning. London: Her Majesty's Stationery Office, 1991.

Mactaggart, P. "Examination and restoration of paint on musical instruments." Restoration of Early Musical Instruments. Occasional Papers No. 6. London: United Kingdom Institute for Conservation, 1987, pp.6-8.

Mann, P.R. "Working Exhibits and the Destruction of Evidence in the Science Museum." International Journal of Museum Management and Curatorship, 8 (1989), pp.369-387.

Michalski, S. "Relative Humidity in Museums, Galleries, and Archives: Specification and Control." Bugs, Mold & Rot II. Washington D.C.: National Institute of Building Sciences, 1993.

Mills, J.S., and White, R. The Organic Chemistry of Museum Objects. London: Butterworths, 1987.

Museums & Galleries Commission. Adhesives. Science for Conservators Vol. 3. London: Routledge/Museums & Galleries Commission, 1992.

Museums & Galleries Commission. An Introduction to Materials. Science for Conservators Vol. 1. London: Routledge/Museums & Galleries Commission, 1992.

Museums & Galleries Commission. Cleaning. Science for Conservators Vol. 2. London: Routledge/Museums & Galleries Commission, 1992.

Museums & Galleries Commission. Managing Your Museum Environment. London: Her Majesty's Stationery Office, 1994.

Museums & Galleries Commission. Standards in the Museum Care of Musical Instruments. London: Museums & Galleries Commission, 1995.

Museum Documentation Association. SPECTRUM: The UK Museum Documentation Standard. Cambridge: Museum Documentation Association, 1994.

Myers, A. "Cataloguing standards for instrument collections." CIMCIM Newsletter, XIV (1989), pp.14-28.

Myers, A. Historic Musical Instruments in the Edinburgh University Collection: Catalogue... Volume 1: The Illustrations. Edinburgh: Edinburgh University Collection of Historic Musical Instruments, 1990.

Myers, A. Historic Musical Instruments in the Edinburgh University Collection: Catalogue... Volume 2: The Descriptions. Edinburgh: Edinburgh University Collection of Historic Musical Instruments, 1992-.

Myers, H., ed. Ethnomusicology: An Introduction. London: Macmillan, 1992.

Newton, R. "Deterioration of glass." R. Newton and S. Davison, eds. Conservation of Glass. London: Butterworths, 1989.

O'Brien, G. "The conservation of historical keyboard instruments: to play or to preserve?" Per una carta Europea del restauro. E.F. Barassi, and M. Laini, eds. Florence: Società Italiana di Musicologia, Leo S. Olschki, 1987, pp.291-297.

Oddy, A., ed. The Art of the Conservator. Washington D.C.: Smithsonian Institution Press, 1992.

Odell, J.S. "Musical Instruments." Caring for Your Collections. New York: Harry N. Abrams, 1992, pp.128-137.

Orna, E., and Pettitt, C. Information Handling in Museums. London: Bingley, 1980.

Orna, E. Information Policies for Museums. Cambridge: Museum Documentation Association, 1987.

Pearson, C. "Code of Ethics and Guidance for Conservation Practice." ICCM Bulletin, 8, 1 & 2 (1982).

Perret, D. "Fiche organologique CIMCIM." CIMCIM (IAMIC) Newsletter, III & IV (1975-76), pp.29-34.

Pinniger, D. Insect Pests in Museums. London: Archetype Publications, 1994.

Prytulak, G. Threaded Fasteners in Metal Artifacts. Technical Bulletin 17, Ottawa: Canadian Conservation Institute, 1997.

Roberts, D.A. Planning the Documentation of Museum Collections. Duxford: Museum Documentation Association, 1985.

Rogers, B.A. The Nature of Metals. Cambridge, MA: The M.I.T. Press, 1964.

Rose, C. "A code of ethics for registrars." Museum News, 63, 3 (1985), pp.42-46.

Sadie, S., ed. The New Grove Dictionary of Musical Instruments. London: Macmillam, 1984.

Sandwith, Hermione, and Stainton, Sheila. National Trust Manual of Housekeeping. London: Viking/National Trust, 1991.

Tétreault J., and Williams, S. Guidelines for Selecting Materials for Exhibit, Storage and Transportation. Ottawa: Canadian Conservation Institute, 1993.

Thomson, G. The Museum Environment. 2nd edition. London: Butterworths, 1986.

United Kingdom Institute for Conservation. "Guidance for Conservation Practice." ICCM Bulletin, 8, 1 & 2 (1982), pp.11-14.

U.S.A. National Standards Institute. American Standard Acoustical Terminology. S1, 1-1960. New York: ANSI, 1960.

Von Hornbostel, E.M., and Sachs, C. "Systematik der Musikinstrumente: ein Versuch." Zeitschrift für Ethnologie, 4 & 5 (1914).

Von Hornbostel, E.M., and Sachs, C. "Classification of Musical Instruments." Trans. A. Baines and K.P. Wachsmann. Galpin Society Journal, XIV (1961), pp.3-29.

Waitzman, M. "Care of... Regular maintenance of your keyboard." Harpsichord and Fortepiano, 5, 1 (1994), pp.41-42; 5, 2 (1995), pp.40-41; 5, 3 (1995), pp.38-39.

Waitzman, M. "From ‘Ancient Musicland' to ‘Authenticity.'" Music and Musicians, 3, 37 (1988), p. 22.

Watson, J.R. "Historical Musical Instruments: A Claim to Use, An Obligation to Preserve." Journal of the American Musical Instrument Society, XVII (1991), pp.69-82.

White, J.R. Jr. "Facadism: Is This Really Preservation?" Locomotive and Railway Preservation, 15 (1988), pp.33-40.

Wraight, D. "The Conservation of Keyboard Instruments." Per una carta Europea del restauro. E.F. Barassi, and M. Laini, eds. Florence: Società Italiana di Musicologia, Leo S. Olschki, 1987, pp.299-303.

Young, R.W. "Pitch notation..." Journal of the Acoustical Society of America, 11 (1939) pp.134-139.

Return to Contents

Link to CIMCIM Welcome Page

Address for further information

This page updated: 30.6.05