MANAJEMEN PRESERVASI INFORMASI

There is really not much scope for aiding the preservation of paper by imposing a low relative humidity. A sheet of paper may be dried without catastrophic effect, but books react differently. They are made of different materials laminated together. As the relative humidity is lowered, these materials change their physical properties in different ways. Paper and leather will shrink and stiffen. Paper will shrink differently in different directions. Cloth, on the other hand, will expand in area as the relative humidity diminishes. Glue becomes extremely hard and brittle as it dries. These materials working againts one another cause book covers to warp. These effects are not yet documented by reliable published data. Apparently a relative humidity below about 35 per cent is rather risky. The upper limit for good conservation is set by the danger of mold growth at about 70 per cent RH. Within this range the deterioration rate of paper seems to vary by a factor of between 2 and 4. The damage done by handling dry and stiff, rather than damp and pliable, paper is very difficult to quantify. At high relative humidity the various materials continue to expand at different rates, but they are much more forgiving to each other and will creep and deform to reduce the stresses caused by movement. I believe that within the limits of 40-60 per cent relative humidity the choice should be controlled by the local climate and by the nature of the outer wall of the building, as I will explaind later. It is better to have a constant relative humidity within this range than to strive towards an ideal relative humidity which can only be obtained in some seasons of the year. Different parts of the building can have different levels of relative humidity. Buildings that are vulnerable to winter condensation should have offices around the perimeter in which a low relative humidity, down to 25 per cent perhaps, is maintained in winter, with the collection confined to an inner region with higher relative humidity. It is not advisable to move books from air at one relative humidity to another environment that is more than 5 per cent different in relative humidity because changing the equilibrium moisture content of a book is a very slow affair indeed. It takes weeks, rather than the hours needed for temperature equilibrium. If books have to be subjected to sudden changes of relative humidity because of the way the library operates, it is much better to have a rather high general level, around 60 per cent, because climate changes at high relative humidity are much less traumatic for the book. I strongly recommend a uniform relative humidity throughout the stack and reading area. The natural tendency of the air-conditioning systems, as of the outside world from hour to hour, is to operate at a constant dew point. This means that in a single duct system every room must be at the same temperature to be at the same relative humidity, unless special provision is made by installing a secondary humidifier in the branch duct leading to a room. This allows a room to be at a higher temperatur. Such local humidifiers are often inserted because the specification calls for very tight limits on relative humidity variation. These devices, so close to the room, can cause rapid oscillation of relative humidity about the set value, which is surely as harmful as an occasional slower wandering from the specified value. Such a local humidifier can cause trouble in another way : if the room humidistat calls for more humidity, the steam injector in the duct will add moisture. The dew point of the room air will increase and may exceed the temperature of the duct, which is carrying cool air. Condensation on the outer surface of the duct will drip into the room. It is important that the curator understand the consequences of setting close tolerances on the climatic variation allowed. A less onerous specification, combined with some of the passive climate control methods discussed later, may well give a more reliable total system. As a postscript to these separate discussions on the effects of temperature and of relative humidity, a comparison of the preservative effects of low temperature on the one hand and of low relative humidity on the other is of interest. The dew point attainable in a building is dependent on the technology of the device used to cool the air. Once a design dew point has been defined, one has a certain freedom in the choice of temperature, or of relative humidity, but not of both independently. For example, air at a temperature of 220C and 33 per cent RH has the same 50C dew point as air at 120C and 62 per cent RH. The damage done by the high temperature of the first set of conditions is offset by the preservative effect of the low relative humidity. It seems from the slender evidence available that the low temperature-high relative humidity alternative gives better preservation.[11] People have a great tolerance for low relative humidity with little tolerance for low temperature, while books have a great tolerance for low temperature with little tolerance for low relative humidity. For a design dew point of 50C a good compromise would be a room temperature of 180C with a relative himidity near 42 per cent. Cold storage vaults in the same building should be held slightly below 42 per cent RH, depending on the temperature, so the moisture content of the paper does not change when a book is withdrawn and warmed up to the reading room temperature. Effect of the Indoor Climate on the Building Envelope Let us turn now to the extremely important but rather neglected subject of the influence of atmospheric moisture on the building. [12] The problem can be simply stated, but not so easily solved. If the air inside the building has a higher dew point than the outside temperature, and if it diffuses or flows through the wall, then somewhere within the wall condensation will occur. The process is exactly tha same as condensation on windows, but since it is invisable, it is presumed not to happen, or at least not to matter. In fact, the window condensation is sometimes the less harmful event. Moisture in walls, and in roofs, can cause serious damage very quickly. Old buildings are vulnerable because they have no provision for preventing the diffusion of air through walls. New buildings are vulnerable because they are well insulated so that the interior surface of the outer skin of the wall is close to the outside temperature. The results of condensation can be increase corrosion of metal fastenings within the wall; movement of soluble salts, whose recrystallization physically disrupts the masonry; and frost damage from the expansion of ice lenses within the wall. In continental climates, these conditions pose a serious threat to buildings. Vapor barriers are cutomarily installed to prevent diffusion of water vapor through the wall. They seldom work as intended, because they are rarely installed carefully enough. The barrier must be airtight. If it is not, it will allow some air through to deposit dew in the wall, and then it will inhibit the free air circulation that would, in warmer weather, evaporate the water. Old buildings are generally not worth making airtight. A thorough survey can give reason for optimism, however. Thick, porous, salt-free masonry walls are not unduly vulnerable if they do not have iron cramps within them. Some old buildings faced with porous stone have an impermeable backing to the stones originally put there to prevent salt migration from brick work. I am not suggesting, therefore, that old buildings can never be humidified to counter the desiccating effect of warmed winter air, but it must be done cautiously, and after a through structural survey. One welcome side effect of running a building cool is that this danger of condensation in the walls is much reduced. Not only is the dew point lower, but the amount of water vapor carried by a given volume of air diminishes sharply with temperature so that the damage done by cool humid air diffusing out to a very cold outside wall surface is very much less than the damage done by warmer air diffusing out to a surface that is an equal number of degress below its dew point. As I mentioned before, air-conditioned buildings are customarily operated at a slightly raised pressure so that any leakage will be of conditioned air outwards rather than of raw outside air inwards. This pressure difference is designed into the system; it is seldom imposed by active measurement and control, although there is no reason why this cannot be done. It is advantageous to run a building which is an a cold climate and is humidified to more than 35 per cent in the winter at, or slightly below, atmospheric pressure. This procedure will not solve the condensation problem entirely, because it is quite possible for air to be entering the building from outside in one pleace, but for another part of the wall to be exposed to inside air. This phenomenon is very common in buildings with high open interior spaces such as domes. The warm, moist air within the building is less dense than the cold dry outside air and so it tends to rise, escaping from the roof while outside air enters down below. Another phenomenon operates in buildings with cavity walls. The general air flow may be inwards, but the air entering the wall will move sideways in the cavity towards fissures in the inner wall. Water vapor diffuses through the pores in the inner wall and crosses this cavity air stream diagonally to condense on the outer leaf of the wall. In summer a danger is the condensation of moist, warm outside air near the inner surface of the wall. This air may be moister than you might suppose from the local wcather report because an additional burden of water vapor can be coming from evaporation of water that has condensed during the winter. The evaporation rate increases dramatically at the high temperature reached within a sunny wall or roof on a warm day. This spring re-activation of water accumulated over the winter can produce spectacular flows of condensate. A building that is not airtight and is in a place with cold winters and hot summers should be operated at a positive pressure during the summer and at a negative pressure during the winter. This seasonal change in operating pressure of air-conditioning systems is never deliberately used, as far as I know, but it is undoubtedly helpful to the preservation of the building. I discovered this when modifications to the air-conditioning system of a museum building unexpectedly altered the direction of air flow and dramatically reduced the winter condensation within the walls. The relative humidity and temperature limits are thus set by the method of construction of the building and by its local climate. There is no easy way to predict the performance of an individual building., nor is it easy to measure the direction and vigor of air movement within the walls and roof. Buildings that are vulnerable to damage by condensation should have sensors buried in the walls and roof to record the state of the climate within the structure. Only a few studies have been published on this subject,[13] but the growing practice of humidifying museums and libraries, and modern trends in insulation for energy saving, no doubt combine to cause damage to buildings in cold climates. In continental climates the best argument for maintaining a low winter relative humidity is the preservation of the building rather than the preservation of the paper within it, although one must not forget the damage done by low relative humidity to furniture and panelling. In humid summer weather the burden of maintaining this constant low relative humidity becomes very expensive. It seems reasonable, therefore, to allow the relative humidity to change slowly through the seasons between limits that should not exceed 35-62 per cent. Books take a very long time to re-equilibrate to a different relative humidity and the stresses imposed by the transient uneven moisture content are not beneficial, so a very slow seasonal change should be imposed. Passive Climate Control We have no doubt that books should, if possible, be kept in a constant climate. Much can be done, however, to safeguard collections without the benefit of the mechanical and electronic systems that I have described. Fortunately, books and documents are rather easy to care for in less than ideal climates. The first basic principle is to put them in a set of nearly airtight and close-fitting enclosures. The second principle is to prevent rapid temperature change around the collection. The effect of enclosure, apart from its obvious role in excluding dust, needs some explanation. The physical properties of paper, such as its dimensions and its stiffness, depend on the moisture content, which is usually around 6 per cent. This water is loosely bound and will be lost to surrounding air of low relative humidity. More water will be absorbed if the surrounding air has a high relative humidity. If the air surrounding the book is isolated from the rest of the atmosphere and if this air volume is kept small compared with the volume of the book, then this exchange of water will be very small, because, although the relative humidity of an isolated volume of air will change with temperature, the amount of water that has to move from book to air, or the other way, to maintain the equilibrium is entirely negligible. The book regulates the moisture in the air trapped around it.[14] Some complicated movements of water can occur on a small scale if a temperature gradient exists across the enclosure. Good thermal buffering is important and can be achieved by insulation, by close stacking of the books, by massive contruction of the building, and by keepng the book stacks away from outside walls, or the walls of furnace rooms. A slow change in air temperature that causes no steep temperature gradients around a book does no harm. A sudden upward temperature change is not too damaging either. If a book in a container is suddenly cooled, condensation will occur when the walls of the container drop below the dew point of the air within. The book will release to the air more water vapor to compensate for that lost, and a rather large amount of water will condense and drip on the book or flood the bottom of the enclosure. From there it will re-evaporate, setting up a cyclic process that will stain the book. This effect is well demonstrated by the glass-fronted boxes fixed to the walls of restaurants. If the menu is not changed frequently in winter, the passing gourmet will soon notice a stain creeping upward past the dessert list. A very similar affliction can damage books in cases againts uninsulated outer walls. The cold of a winter night can cool the back of case, while the books remain warm because the glass front readily transmits the heat of the room. The books maintain the relative humidity of the air in the container. This condition is normally a virtue, but now it becomes a source of danger, as the condensed water dribbles invisibly down the back of the bookcase. Less dramatically, but more often, the relative humidity close to the cold back will only rise to some value above 70 per cent, which allows dormant fungal spores and filaments to become active. It is dangerous to allow the temperature of any part of a book stack to fall much below the temperature of the main body of air in the room because a roomful of air quickly achieves a uniform moisture content by convective mixing. If the temperature differs from place to place, the relative humidity must vary also. Cold walls have a boundary layer of cold air close to them which does not so readily mix into the general air circulation in the room. If water is rissing in the wall, or penetrating through a porous wall, the moisture content of the air may locally be high. The two independent effects combine to give a dangerously high local relative humidity. The intermediate technology solution to this situation is to scour away the boundary layer with a draft from a fan. The permanent solution is to put insulation on the wall and cover this with an impermeable membrane. If the average climate outside the library is within the limits set for good conservation, that is, less than 25oC and between 40 per cent and 65 per cent RH, then no great harm will come to collections that are not air-conditioned. Some seasons of the year, however, may be beyond these limits. The buffering capacity of the passive climate control measure described above may be exceeded, and the climate within the containers will drift towards a dangerous condition.[15] Air-conditioning, even of a simple kind, then becontes essential. It may be that small free-standing humidifiers or dehumidifiers will cope with these brief seasonal periods of danger. Much can also be done by adjusting the environment near the building. Pale paint reduces heat gain in the sun. trees will do the same, eventually. Grass surroundings will reflect less solar energy onto the façade than will a marble concourse. The adjustment of the microenvironment, which is one of the more enjoyable fantasies of the ecological movement, can tip the balance for buildings that are not in extreme climates. For libraries not yet built, we can rethink the whole concept of library design. On the whole, the custom of building massive, prestigous shrines to learning has served the cause of conservation well. Their natural temperature stability leads to good relative humidity stability because the daily fluctuation in atmospheric relative humidity is mainly caused by the daily temperature cycle. The moisture content of the air changes less often since it is controlled mainly by the direction of origin of the air mass that covers the region. More radical solutions are available, however, such as partial, or nearly complete, burial of the building. Existing buildings can be adjusted, for example, by installing a ventilated attic space to reduce heat conduction through the roof and to prevent roof condensation in a humidified building. Finally, in this varied collection of measures to modify the indoor climate, one should not neglect the pleasure that can be obtained from air-conditioning devices. An ornamental fountain in the lobby can be used to humidify, or to dehumidify the air, according to the water temperature. Air Pollution Originating within the Building Much has been written lately on the subject of air pollution generated indoors.[16] In libraries the important pollutants are formaldehyde, formic acid, and acetic acid emitted by wood, particularly plywood and particle board. Some humidifiers release gases into the air, such as diethylaminoethanol, which are used to inhibit the corrosion of steam pipes. This chemical is a hygroscopic alkaline vapor which probably reacts with the acid pollutants in air to form nonvolatile salts, which precipitate as a slimy film on surfaces. One unusual hazard in libraries is the oxides of nitrogen released by the pyroxylin cloth used to cover books. Damaging chemicals are also released within books, but climate control cannot cure this problem because the molecule will react long before if can diffuse out from the book. We are concerned, therefore, about chemicals that emerge from the container and from the outside surface of the book. Wood should really not be used in libraries, but, of course, it will continue to be used and to survive, so the collection must be protected against its outgassing. Various techniques can be used, depending on the circumstances, fierce ventilation is the traditional, but not now popular, method. Some research is in progress on the efficiency of various surface coatings. Until results of these studies are released, a layer of aluminium foil is recommended as an impermeable barrier, or a layer of paper impregnated with calcium carbonate, which will react with the mainly acid gases emitted from wood. This paper should also catalyze the transformation of formaldehyde into formic acid and then react with the product, but research has not yet demonstrated that this reaction actually takes place. If the container is lined with foil or alkaline paper in this way, it seems that tight containment of books presents less danger than exposing them to the room air. It certainly ensures a constant climate. For those who are not convinced by these arguments, even a permeable container such as an ordinary envelope slows down fluctuations in relative humidity, reacts with pollutants, and gives useful protection to the contents. Conclusion The influences on paper storage of climate, building methods, air-conditioning technology, pollution chemistry, and reader comfort are varied and interwined. In such a complicated environment reliance on an arbitrary set of standards can lead to great expense and ultimate failure if the standards prove impossible to realize in a particular geographical and social environment. Each institution must be regarded as a unique system for which a unique compromise must be developed by intelligent study. A knowledge of the principles which underline the standards and codes of practice that have been developed over the years is essential. Notes 1. A good general text on the effects of climate on historic materials is : Thomson, G. The museum environment. London : Butterworth, 1986. 2. The standard reference text for air-conditioning design in North America is the series of handbooks produced by the American Society of Heating, Ventilating and Air Conditioning Engineers, of Atlanta, Georgia, USA. In particular the Handbook of Fundamentals covers many of the subjects discussed in this article. 3. A standard work on the interaction of cellulosic material and water is : Hearle, J.W.S., and Peters, R.H. Moisture in textiles. New York: Textile Book Publishers, 1960. 4. A useful introduction to air-conditioning technology is : Shuttleworth, R. Mechanical and electrical systems for construction. New York : McGraw-Hill, 1983. 5. Methods of pollution control are discussed in : Mathey, R.G., Faison, T.K., and Silberstein, S., Air quality criteria for storage of paper-based archival records. NBSIR-83-2795. Washington : National Bureau of Standards, 1983. See also : Padfield, T., Erhardt, D., and Hopwood, W. 'Trouble in store'. In International Institute for Conservation of Historic and Artistic Works. Science and Technology in the Service of Conservation, Preprints of the Contributions to the Washington Congress, 3-9 September 1982. London : International Institute for Conservation of Historic and Artistic Works, c1982, 24-27; Indoor pollutants. Washington, D.C.: National Academy Press, 1981. 6. Relative humidity measurement is reviewed in : Moisture and humidity 1985. Research Triangle Park, North Carolina : Instument Society of America, 1985. 7. The N.B.S. report, reference 5, proposes three grades of storage : 18oC-24oC and 40-45 per cent RH for immediately accessible books, 10oC-13oC and 35 per cent RH for less frequently used materials, and -29oC for cold storage. An American National Standard, Z39.54-198X, is currently being prepared. A committee of the U.S. National Academy of Sciences is also deliberating on the subject. There is a British Standard, BS 5454:1977, Recommendations for the Storage and Exhibition of Archival Documents. 8. Graminski, E.L., Parks, E.J., and Toth, E.H. ' The effects of temperature and moisture on the accelerated aging of paper'. In Durability of macromoleculer materials. Washington, D.C.: American Chemical Society, 1979, 341-355. (ACS Symposium series no.95) 9. Nanassy, A.J. 'Temperature dependence of NMR measurement on moisture in wood'. Wood science, October 1978, 11(86). 10. Graminski, Parks and Toth, op.cit. 11. Ibid 12. See the ASHRAE Handbook of fundamentals (reference 2) and also Lieff, M., Trechsel, H.R. Moisture migration in buildings. American Society for Testing and Materials, 1982. (special Publication779) 13. Moisture and humidity 1985, op.cit. ; ASHRAE Handbook of fundamentals, op.cit. ; Lief and Trechsel, op.cit. 14. Padfield, T., Burke, M.,and Erhardt, D.A. cooled display case for George Washington's Commision'. In International Council of Museums, 7th Triennial Meeting, Copenhagen, 10-14 September, 1984. Preprints. France : ICOM, 1984. 15. Thomson, op.cit. 16. Padfield, T., Erhardt, D., and Hopwood, W. 'Trouble in store'. In International Institute for Conservation of Historic and Artistic Works. Science and Technology in the Service of Conservation. Preprints of the Contributions to the Washington Congress, 3-9 September 1982. London : International Institute for Conservation of Historic and Artistic Works, c1982, 24-27; Indoor pollutants. Washington, D.C.: National Academy Press, 1981. Selections from Parker, T.A. 'Integrated Pest Management for Libraries,' in Preservation of Library Materials : Conference Held at the National Library of Austria, Vienna, April 7-10, 1986, edited by Merrily Smith. IFLA Publications 40-41. Munchen :K.G. Saur, 1987), Vol.2, pp.103-123.[1] This paper was first published in Preservation of Library Materials, IFLA Publications 40-41 and has been republished with the permission of IFLA. Some material is omitted. The reader should refer to the original for illustrations and description of carpet beetles, cigarette beetles, drugstore beetles and psocids and for information about methods to control these pests. Also omitted is materials about some fumigation techniques. Abstract A library is, in effect, a concentration of foodstuffs for the common pests-insects, rodents, and mold-that attack the collection. The best control of pests is furnished by an integrated pest program, i.e., the use of a combination of control techniques. Insect damage to library materials is caused primarily by cockroaches (mainly the American, Oriental, and Australian cockroaches); silverfish (thirteen species are known in the Unitd States); carpet beetles (the larvae of several species are damaging); cigarette beetles (the most common pest of herbarium collections); the drugstore beetle (sometimes called a 'bookworm'); and psocids, or book lice. The common rodent found in libraries, particularly in subtropical and tropical climates. All of these infestations can be dealt with by the concerted use of various techniques, including external precautions to buildings, insect traps, the use of insecticides and other chemicals, control of moisture, cleanliness measures, limited heat treatment of infested materials, and continual inspection for evidence of infestations. Fumigation of library materials may be warranted in some instances, but is rarely necessary. A continual awareness of potential problems and immediate treatment are essential. Man has come to realize that to one approach to pest prevention and control will suffice. Instead a combination of techniques is usually required to maximize the effectiveness of any pest control program. The term 'integrated pest management' (IPM) has been coined to embody this concept : that all pest control programs must rely on several approaches working in concert to effect the desired result. An IPM approach must be considered when addressing the problems of pests in libraries. A library, where books, printed materials, manuscripts, maps, prints, photographs, and archival materials are stored, perused, and exhibited is not unlike the setting in agriculture where huge quantities of foodstuffs are stored for long periods of time. The library is a concentration of foodstuffs, including starches, cellulose, and proteins, which forms a banquet for insects, rodents, and mold. In addition, the environment in which these foodstuffs are stored is indoors, protected from extremes of harsh climates. Populations of insects specific to this micro-environment can easily explode and cause serious damage if IPM approaches are not utilized fully to prevent such an occurrence… The most common pests encountered in libraries are insects, rodents, and mold. Each of these problems will be reviewed, and the IPM approaches necessary for their control and prevention will be outlined. Insects Damage to library materials from insects is primarily caused by cockroaches, silverfish, various beetles, and book lice. Damage to these materials results when insects use them as a food source. Both immature and adult stages of cokroaches, silverfish, and book lice cause feeding damage on library materials. In the case of beetles, it is primarily the larval stage that is responsible for the feeding damage. The larvae chew their way through a book, ingest the material, and leave a tunnel filled with powdered excrement. Once the larvae have completed their development, they pupate, and the adult beetles emerge by chewing their way out. Small round exit holes are left in the book. Cockroaches Substantial damage to library materials can be attributed to various large species of cockroaches. These problems are more prevalent in the subtropical and tropical areas of the world, but damage can also be found in temperate climates. Three cockroaches in particular are notable for the damage they do to library materials. They are the American cockroach, Periplaneta americana (Linn.); the Oriental cockroach, Blatta orientalis (Linn.); and the Australian cockroach, Periplaneta australasiae (Fabr.). These cockroaches have large, strong, chewing mouth parts, prefer starchy materials, and can easily destroy paper, paper products, bindings, and other coverings on books and pamphlets. Chewing damage is generally recognized by smears of fecal material in association with the damage and a ragged appearance to the areas that have been fed upon. These areas generally appear around the edges of the piece where small bits of paper have been removed and eaten. Sometimes pelletized droppings are also found in association with the feeding, particularly with the American cockroach…. The large species of cockroaches can be controlled by the following IPM measure. 1. Installation of a gravel 6-fot barrier around the perimeter of the library to prevent ingress from outdoors 2. Elimination of all vines and ivy from the building. 3. Installation of proper screening on all windows and doors. 4. Installation of exterior lights away from the building so they will shine on the building from a distance rather than fastening them to the building so that insects are attracted to the exterior walls during the night. 5. Removal of all debris, leaves, and twigs around the exterior of the library as well as cleaning out debris from gutters on the roof of the building. 6. Elimination of cockroach harborages and entries by caulking and sealing. 7. Installation of sticky glueboards that will trap insects on their nightly forays around the library. These insects traps can be installed in false ceilings, basement, elevator shaftways, and closets so they intercept insects as they travel looking for food. 8. Whenever insect infestations are found, a common attempt at their control is by insecticidal treatments with an aerosol or fog. Insecticidal fogs or aerosols should never be used in any collection. Such formulations are normally oil-based. During application, small droplets of the oil/insecticide mixture are dispended into the air, eventually settling on the entire collection. This kind of treatment irreversibly damages the collection. 9. The use of insect baits, such as 2 per cent Baygon Cokroach Bait, applied sparingly to quiet zones of the interior of the library. This bait, which looks like sawdust, is bran mixed with molasses and contains 2 per cent Baygon. This bait is a favorite of large cockroaches and will easily bring populations into control. 10. Perimeter fan spraying with residual insecticides, paying particular attention to those areas adjacent to pipe chases, elevator shafts, storage areas, and mechanical rooms. 11. In the case of the American and Australian cockroaches, perimeter, exterior power-spraying of the walls and overhangs may be required. 12. Installation of thresholds and rubber flaps on exterior doors to prevent ingress by cockroaches from the exterior, particularly at night. 13. the used of steel wool in holes and openings leading from drainage and sewer systems to prevent cockroach ingress. Silverfish Silverfish…are one of most common pests of libraries. All have weak, chewing-type mouth parts and tend to feed on products high in carbohydrates (starch) and proteins. Such materials as paper, paper sizing, prints, glue and paste, wallpaper, and drywall are favorites of silverfish. Damage from the feeding of silverfish can be recognized by certain areas that have been eaten all the way through and other areas that have only partially been eaten through. Silverfish tend to rasp their way slowly through a piece of paper, and it is this damage that is seen on library materials and prints. Silverfish will roam widely in search of foods, but once they have found a satisfactory source, they remain close to it. There are many species of silverfish in the world; thirteen are known from the United States. Some prefer cool, moist environments, others warm and moist environments. They are small, tapered, wingless insects with long antennae and three long bristles protruding from their posterior end. They are nocturnal, resting in cracks and crevices during the daytime. When exposed to light, they move quickly to avoid it. Silverfish are particularly fond of paper with a glaze on it. Often sizing, which may consist of starch, dextrin, casein, gum, and glue, is particularly attacked. Certain dyes are attractive to silverfish. Studies have shown that papers consisting of pure chemical pulp are more likely to be attacked than those consisting in part of mechanical pulp. In general, papers and books in regular use are not damage by silverfish. Silverfish are also particularly fond of rayon and cellophane. In temperate climates, silverfish tend to migrate vertically depending on the season of the year. In the hot months of summer, silverfish will migrate down into the cooler, more moist portions of the building, and in the fall and winter they will tend to migrate to attics and higher levels. Drying out a building with heat in the winter time will help to reduce silverfish populations. The heat also eliminates the microscopic mold that grows on plaster walls and drywall providing a food source for silverfish. In cool, moist basements, and commonly in poured-concrete buildings, silverfish are a year-round problem. It is impossible to eliminate bringing silverfish into a library. Silverfish are a very common problem in cardboard box and drywall manufacturing facilities silverfish lay eggs in the corrugations of cardboard boxes, one of their favorite areas for egg deposition. With every cardboard box coming into a library, a new load of silverfish and their eggs is bound to arrive. Upon hatching, silverfish go through many molts throughout their lifetime and have a long life span. Control and prevention of silverfish damage to library materials can be effected in a variety of ways. 1. Thorough vacuuming of the perimeters of rooms where silverfish like to hide underneath the toe moldings and baseboards during the daytime. 2. The use of insect sticky traps or glueboards in those areas where silverfish seem to be a problem. Each night, when silverfish are active, they will be trapped on the glueboards. 3. In cabinet storage situations, the use of silica gel in a finely powdered form in the void space beneath the bottom shelf or drawer of a cabinet. Sometimes the use of a 1/4-inch drill is necessary to gain access to apply this dust into the voids beneath the cabinets. Silica gel is a desiccant and kills silverfish by drying them out. By placing silica gel powder (sometimes in combination with pyrethrum insecticide) beneath the cabinets, there is virtually no way a silverfish can crawl up into the cabinet without encountering the silica gel thus being repelled or killed. 4. In some instances, particularly in manuscript, rare book, and print collections, the use of insecticide resin strips may be used in enclosed spaces. Insecticide resin strips contain the insecticide Vapona (DDVP). This material volatilizes from the resin strip and fills a confined space with molecules of insecticide. It is a mild fumigant and in time will kill all stages of insects within the enclosed space. The normal rate of application of these strips is one strip per thousand cubic feet of enclosed space.[12] This type of chemical application is for enclosed spaces only, such as cabinets, vaults, and small storage rooms, and is not designed to be used in open, public spaces or where ventilation would carry the fumes out of the space. 5. Application of residual, liquid insecticidal sprays to perimeters of rooms and at the base of all stack shelving areas, paying particular attention to that crevice where the floor meets the wall, baseboards, and shelving. 6. The application of insecticidal dusts, such as a silica gel/pyrethrum combination, to voids where pipes enter wails and penetrate floors and in wall voids. In warmer parts of the world, construction engineers sometimes inject powdered silica gel in wall voids during construction of a building. This practice eliminates insect harborages. 7. Crack and crevice injection of small spot applications of liquid insecticides to the backs of cabinets where they are attached to walls. 8. The control and elimination of moisture such as leaky plumbing, around laundry areas, in bathrooms, and workrooms where a silverfish population can thrive because of the high moisture situation. 9. Reduction of potential sites of harborage by the use of caulking compounds and patching plasters…. Rodents The most common rodent found in libraries is the House Mouse, Mus musculus….This species seems to be able to invade practically any structure man has made. Damage to library materials comes from mice destroying materials for nesting purposes, and urinating and defecating on library materials. Populations of mice can build up very quickly, and when they die, their carcasses act as a source of food for carpet beetles. In addition to damaging collections directly, mice may chew the insulation off electrical wires, causing them to short and start a fire. House mice are secretive and are generally active at night. They live in a territory with a small home range. The average distance that a mouse travels in its activities is 12 feet. Male mice are highly territorial, and for this reason control measures must be designed for specific areas where mouse droppings re found. Glueboards and traps should be set in these areas. House mice live outdoors year around, but will invade buildings, particularly in the fall of the year in temperate climates…. Mice feed on a variety of foods provided by man. They also feed on dead insects found indoors. Mice have been found to be cannibalistic. They apparently do not need free water to drink, but will consume it if it is available. Mice feeding on high protein diets must supplement their diets with free liquid. During their nocturnal activity, mice leave fecal droppings wherever they have been active. Other signs of mouse infestation are gnaw marks; small, stained holes in floors and walls, and beneath doors; and a pungent odor from their urine. Mouse control in a library is important and should be dealt with in the following ways. 1. Seal the building on the exterior as tightly as possible with steel wool and caulking compounds. 2. Never use a toxic baiting program for mice on the interior of a library. The mice will die in the walls, floors, and ceilings, and provide food for carpet beetles. 3. Use mechanical control techniques for mouse control. Snap traps baited with cotton balls or peanut butter can be used to trap mice. Multiple-catch live traps are available on the market, such as the 'Ketch-all' trap. These are capable of catching more than one mouse at a time without the use of bait, relying on the innate curiosity of the mouse. Glueboards can also be used to trap mice. Soon after they are trapped they will die, and the trap or glue board can be discarded. 4. A thorough inspection of the building with a flashlight on a periodic basis is important to identify those areas where mice activity is present. The presence of droppings is a clear indication that control measures should be undertaken in that exact spot. Two or three weeks after a control program has been instituted, remove all droppings so the progress of the trapping program can be determined. 5. In temperate regions, in late summer and early fall, trapping programs should be instituted so they will be in place when the mice naturally tend to invade structures. 6. Sonic devices aimed at rodent elimination are of questionable value in most library mouse control programs. Mold and Mildew One large problem in libraries, particularly in subtropical and tropical climates, is the presence of mold on libraray materials. Mold results from spores landing on a substrate that has the correct temperature and surrounding humidity to initiate germination of the spores. When the spores germinate, they put out fine strands of mycelia, which invade the subsrate, utilizing it as a food source. The mold mycelia exude liquids that dissolve the substrate, and this food is then used in the production of more mycelia and eventually millions of spores. In order for this scenario to take place on paper products, books, and other library materials, prolonged periods of high humidity are required for mold growth. If the environment of a library is held at a temperature of from 680F to 720F and a relative humidity of 50 per cent to 60 per cent, mold will not be seen. This is not to say that some spores will not germinate. It simply means that after germinating, the mycelia will not have conditions suitable for growth and will die before being visible to the naked eye. When the relative humidity of the environment remains in the 60 per cent to 70 per cent range, certain kinds of mold spores will germinate, but most will be unable to maintain my celial growth and will collapse. This higher humidity range is not as 'safe' for paper products, because there will tend to be micro-environments in the library where humidities will peak higher than the overall relative humidity for the entire structure and may create conditions conducive to localized mold growth. Of course, higher temperatures under these higher humidities will also enhance the possibility of localized mold growth on library materials. When the relative humidity of a library exceeds 75 per cent and remains in this range for a period of time, serious mold problems will result on library materials. Even if temperatures are low, the effects of the high humidity will stimulate spores to germinate en masse. Not only will spores germinate, but growth of the resultant mycelia will be quick and unabated. In as short a time as thirty-six hours, mycelial mats will begin to appear on the materials and spread outward. Soon the center of the mat will begin to appear dark, generating millions of spores. The key to long-term mold control, then, is to manage the moisture in the air of the library and stacks in a manner that will minimize periods of high humidity. Each cubic foot of air contains thousands of mold spores which land on surfaces and objects in the library every day. Attempts to control mold on library materials by using various chemicals therefore are usually ineffective. Chemicals such as thymol, ortho phenylphenol (OPP), alcohol, and diluted bleach solutions kill some of the mold spores on the surface as well as some of the mycelia. As soon as these chemicals have volatilized from the surface, the object is vulnerable to new mold spores landing on the surface. If the conditions are correct, germination and production of more mold will result. These types of chemicals do not impart residual control for mold or the mold spores. Similarly, fumigation with poisonous gases in a chamber does not impart any residual mold control effects. Much of the fumigation that is done in libraries is not warranted. Changing the environment that produced the conditions suitable for mold growth in the first place is the only truly effective means of retarding and eliminating mold growth. If a spore lands on a substrate that is not suitable for growth, in time the spore will desiccate and die. As long as the conditions of the substrate and surrounding micro-environment are not suitable for spore germination, the spores will not germinate and mold will never appear. Some considerations in handling mold and mildew in libraries follow. 1. An air handling system should be installed that will lower the humidity in the air and then reheat the air to desired levels. This system should be designed to handle incoming outside air as well as recirculated air. Such systems must be carefully thought out and must be large enough to accept incoming loads with humidity levels of the exterior air as well as the amount of moisture contained in the interior air. The aim is to maintain an interior environment in the library of from 50 to 60 per cent relative humidity and 680 to 720F at all times. 2. If such air handling systems are not available or cannot be installed, fans can be used to keep the air moving, particularly near outside walls and close to floor levels, in an attempt to lower moisture content of library materials. 3. Waterproofing basements and walls below grade on the exterior to prevent moisture from wicking through the walls and into the interior will aid keeping humidity levels down inside the building. 4. Earthen floors in basements and sub-basements should be sealed with concrete to prevent moisture from wicking up into the building. At the very least, earthen floors should be covered with 4-6 mil polyethylene film to lessen the amount of moisture being volatilized into the interior air. 5. Water-sealant paints can be applied to floors and walls to prevent ingress of moisture into the interior of the building. 6. Attick vents and fans can be installed to pull air through buildings that have no air handling systems and where tropical climates require windows to be opened throughout the year. With such installations air can at least be kept moving throughout the building. 7. Open trenches and drains in mechanical rooms and areas adjacent to stack areas should be covered to prevent evaporation of liquid into the interior space. 8. Expcept for drinking fountains, interior fountains or waterfalls should not be permitted in a library. 9. Do not allow indoor planted areas in a library. Keep ornamental and hanging plants to a minimum to reduce the amount of water released into the interior air. 10. Heavy mold infestations resulting from flooding, water damage, leaks, and fires is an entirely separate topic and cannot be dealt with within the scope of this paper. 11. Regular inspection of the collections with a flashlight to pinpoint trouble areas is a necessity. Localized infestations of mold can be temporarily arrested with topical applications of chemicals until other modifications can be made. 12. Thymol is commonly used as a temporary mold-control chemical on books, paper, and other library materials. The use of thymol, either as a mist or spray, or as a fumigant volatilized by heat, does not impart residual mold control to the library materials. Thymol will kill some species of mold spores and mycelia upon contact. Taking the materials out of the atmosphere of thymol will leave them vulnerable to mold spore deposition and possible germination. In the United States, thymol is not registered as a mold-control chemical with the Environmental Protection Agency…. Fumigation Fumigation of library materials with extremely toxic chemicals is rarely necessary. It may be warranted when dealing with bookworms, but fumigation is generally not warranted when dealing with mold and mildew problems. Historically, the library community has used ethylene oxide in fumigation chambers for mold and mildew control on incoming library materials. As was stressed in the section on mold, fumigation will not control mold and mildew if the library materials are placed back into the same conditions from which they came. In most instances library materials that have been fumigated are then stored in areas which do not have an environment conducive to mold growth. The success of the fumigation is given as a reason for the control of the mold and mildew, when, in fact, the new area in which the materials are stored is the governing factor in the mold and mildew control. Ethylene oxide, either in combination with Freon or carbon dioxide, has been found to be a carcinogenic material. In the United States a chamber may have no more than 1 part per million (ppm) of ethylene oxide left after the aeration and before the materials can be removed safely. One of the major problems with ethylene oxide is that very few chambers in the world meet this requirement. Any fumigation chamber relying on an air wash systems to aerate the goods after a fumigation has been completed usually will not reach levels of 1 ppm or below at the end of the air wash cycles…. Ethylene oxide is soluble in oils, fats, and lipids, making leather-bound books retain ethylene oxide for long periods of time after fumigation. After bringing books out of a chamber, they will volatilize ethylene oxide into the air for varying periods of time up to and exceeding three months. It is therefore critical that the managers of major libraries test in-house chambers and study fumigation policies and procedures to determine if they are meeting current requirements. Most will find that fumigation chamber modifications and procedural changes are required to meet current standards. Other fumigants, such as methyl bromide, hydrogen sulfide, and some of the liquid fumigants, are not generally acceptable for library materials for several reasons. Methyl bromide sometimes chemically reacts with materials high in sulfur. If this chemical reaction were to take place, mercaptans would be formed and would create an irreversible, foul-smelling odor. Hydrogen sulfide is explosive and dangerous to use. Some of the liquid fumigants have been found to be carcinogenic. Recently in the United States, Vikane (sulfuryl fluoride), manufactured by Dow Chemical Company, has been registered for use in chambers as a fumigant. As with all fumigants, this material does not impart any residual control, but can effectively penetrate dense materials, such as library materials, and will kill all stages of insects. One problem with Vikane is that it is a poor ovicide, and therefore dosages must be increased in order to penetrate the eggs of certain species of insects. To date, this material has been found to be very nonreactive with materials and is commonly used as a structural fumigant in wood-destroying insect control. Its use to control mold spores and mycelia remains in debate. Conclusion As is true with all museum materials, one cannot delay treatment until insects, rodents, and mold have turned the library collections into a food source. We must be keenly aware of what is happening in our collections and structures. We must anticipate the types of problems unique to libraries and provide an integrated pest management plan to deal immediately with those problems present and to prevent others from arising. In this way, we will establish the most common sense approaches to pest prevention and control with the least impact on our environment and ourselves from toxic chemicals. Notes 1. Dr Parker, an entomologist, is a consultant to museums, historic properties and libraries, specializing in IPM approaches. He is President of Pest Control Services, Inc., 14 East Stratford Avenue, Landsdowne, Pennsylvania, USA, 19050. Telephone (215) 284-6249. 2. The No-Pest Strip is the only dry strip available on the market today. It is available through Kenco Chemical Company, PO Box 6246, Jacksonville, Florida, USA. 32236. Telephone (800) 523-3685. Jackson, Marie. 'Library Security : Facts and Figures.' Library Association Record 93, 6 (June 1991) : 380, 382, 384. First published in Library Association Record 93 (6) June 1991 and reproduced by permission of the author and the Library Association We still do not know enough about the extent of theft in libraries. Statistics are very often estimated losses; some members of the public do not even see 'keeping their books' as a crime. Marie Jackson, National Preservation Officer, pulls together what facts we do have, recounts some horror stories and calls for more research. Given that a fact is 'anything known to have happened or to be true' everyone knows the basic facts of library security. They are : 1. Librarians, if they are fulfilling their task correctly, cannot completely safeguard their collections against theft or abuse; 2. Incidences or theft, non-return and mutilation of library stock are on the increase; 3. Library budgets can no longer (if they ever could) afford the consequences of widespread abuse. It's the figures that are more of a problem and, of course, we are part of a culture which tries to prove with figures what our heads, hearts and eyes tell us is true. We lack firm, proven figures on how many books go missing from libraries in the UK. You have only to look at the NPO (National Preservation Office) bibliography Library Security to see how little published material on library security there is from this side of the Atlantic. A search of LISA, from 1969-87, revealed only 10 British articles. Published figures on book loss are an even greater rarity. What figures are available from libraries on both sides of the Atlantic? Two statistics to start with : in 1894 1 per cent of the population in England used public libraries. At a 1908 conference of the American Library Association, Isabel Ely Lord, Librarian of the Pratt Institute Free Library in Brooklyn, quoted losses in US open-shelf libraries of between seven and thirty-nine books in every 10,000 books circulated. That's a loss rate of 0,3 per cent maximum. From this point on, with massive popularising of the library and huge increases in the number of books held, the modern age takes over. Imagine counting losses for Brooklyn in hundreds these days. Let's not make the mistake, though, of thinking that only twentieth-century man misbehaves in libraries. Chained libraries were there for a reason. The Reverend Charles Burney, the classical scholar whose rare book collection is now revered by British Library users, was expelled from Cambridge after stealing books from the library. It was the Bedmaker who found the proof : 35 classical books in a dark corner which had the University arms removed. If we had firm reason to believe wrong of our users and were allowed to look under beds it might be easier. Before we get near to proving who did it we have to know that something has been done. It is generally true that the statistics that have been gathered on missing books, etc, are estimated losses. They are not the result of detailed shelf and user surveys. Surveys, or inventories, are generally a thing of the past. The reasons are obvious. One of the most detailed UK investigations of publik library abuse was undertaken by CLAIM (the Centre for Library and Information Management) in July and August 1984. a questionnaire to 300 public libraries asked for answers on some 100 issues in all, including incidents of book theft and mutilation. There is no reason to believe that the libraries answering (and 209 did) conducted detailed surveys of the books missing from their shelves. They simply estimated. Eighty-five per cent reported at least one episode of intentional book damage. Forty per cent reported 'chronic' theft of reference material (but 'chronic' for this report, was defined as 'more than six episodes'). Fifty-seven per cent had experienced at least one episode of theft of personal property. 'It is the unusual library that is completely free of crime' the report concludes.[1] An earlier report relevant to academic libraries is the 'Survey of college libraries in the UK 1970' which was published in Library and Information Bulletin, no 15, 1971. this survey, again by questionnaire, convered the whole gamut of college library life, from librarians' qualifications to preservation techniques. Security of collections got only two questions : 'What book losses do you suffer?' and 'What steps do you take to prevent book losses?' The average of actual book loss reported was 2.6 per cent, the estimate 3.09 per cent. Lots of librarians wrote things like 'minimal loss', and 'agreeably small'. '£200 loss of stock each year'. The conclusion of the report is : 'It would appear that the average is reliable but that any individual library would be found to experience losses in a wide bracket; from 0.1 per cent to 10 per cent of actual open shelf stack, though this might vary within the one library from year to year.' The reported methods used to prevent book loss had a top three of :  Continuous manning;  Cases in cloakroom;  Coats in cloakroom. Don Revill wrote an article for New Library World in April 1978 [2] about book loss in academic libraries. In this he says : 'The proportions of the book stock "lost" per annum generally suggest a national average of one per cent. 2.6 per cent was reported for college libraries in the UK by the LA survey of 1970. My own libraries have suffered losses between five and six per cent.' There are all sorts of reasons why estimates of books missing from shelves are not sufficient for a book loss survey. Mis-shelving catalogue or computer error, books on librarins' desks with no record to accompany the 'loan'. They all play a part. A particularly impressive survey was carried out in Long Beach (California) Public Library and Information Center between August 1987and February 1989. Its findings are presented in Library and Archival Security, Vol 10 No.1, 1990. The methodology employed is interesting. Two computer lists were generated – one for books that had not circulated between 1981 and 1987 and one for books that had not been checked out between 1987 and 1988. Two thousand four hundred and fifty-six items which should have been on the shelf were found to be missing. This was 23 per cent of stock. Where were these books ? The possibilities were : 1. a flaw in the security system. Tests on the 3M system were carried out and other libraries were canvassed about their experiences with similar security systems; 2. this particular library had changed its computer. Could some books have been 'lost in the system' during transfer?; 3. human error; mis-shelving and processing; 4. the books had been stolen by (a) patrons, (b) staff. To find the answers, an inventory which included full shelf checking, research in storage and other areas and tests on human error rate in processing was carried out. Patterns were looked for and other libraries canvassed. After all other errors were accounted for (and human error was high), a true loss rate of 8 per cent was reached. As a final corroboration of this figure, staff undertook a survey of the literature on book losses to see how their 8 per cent matched up. They found quoted a loss rate as high as 14 per cent among new titles, a genuine drop in theft after installation in security equipment, that no systems eliminated all loss and that 8 per cent seemed to be a common loss rate. It is interesting that their loss of books from the popular author lists (and remember this is a public library) was as high as 63 per cent though some of this loss was though to be explained by some books having literally been read o pieces, discarded but not noted as discarded. And the four most popular authors in Long Beach? Jackie Collins, Joseph Conrad, William Faulkner and Stephen King. Back to UK figures. In his book Preventing Library Book Theft [3] Herbert Keele quotes (sometimes anonymously) the results of some studies done in various libraries on loss rates. These studies were generally carried out in libraries which had no electronic security system. Indeed, the purpose of the review was often to argue for such a system. In the Metropolitan Borough of Knowsley, a 9 per cent loss rate in four unprotected libraries was noted in 1979. In a large city, four suburban libraries carried out substantial checks to see if a security system could be justified. The checks gave a result ranging from 17 per cent to 40 per cent. On the question of what is most likely to be stolen, Keele argues that equations on loss often neglect the fact that books in some subject areas are more likely to be stolen and new acquisitions are particularly in danger. Keele suggests that, if a 2 per cent loss figure is a assumed, it is probably fair to assume 1.5 per cent loss in new acquisitions, a 0.5 per cent loss of older material. Some other 'facts' and figures which have appeared in published form (though sometimes only in newspapers and, as they misquote me, they probably do the same to everyone else) are :  a questionnaire on library vandalism by a student at Brighton in 1988 shows that 50 per cent of libraries consulted said their service was affected by vandalism;  Lewisham Library apparently shuts when Millwall are playing at home;  Overall, as we would expect, metropolitan libraries had more security problems than rural and suburban ones;  Brent Council, according to press reports, lost 411.000 items over four years. Value £4m. Similar reports quote 20.000 lost books in Liverpool (£100.000). Glasgow lends 6.5m books each year and loses 26.000 (that's of course a very low loss rate if it is correct). What subjects are most likely to be stolen or mutilated? From public libraries, manuals on motor cars and health manuals go missing to a frightening degree. From academic libraries, law publications are seemingly most at risk – understandable given their cost and the pressure on law students to get information quickly. In an article in the Record [4] last year I put forward some estimates of what theft and misuse in libraries is costing us. Given an absolute minimum loss rate of 2 per cent from 150 million books in UK libraries (there are a lot more) and a replacement cost of £20 per volume, a loss figure of £60m is reached. The truth is probably far nearer £100m. The loss rate is almost certainly between 1 and 10 per cent for most libraries. Some libraries without protection systems can reach 30 or 40 per cent. One library quoted in the literature claims to have lost 110 per cent one year : everything it bought that year was stolen; they bought further books and they were stolen. It is vitally important to know the real cost of theft and misuse. Some excellent work done by Andy Stephens at the British Library as part of the Enright review of acquisition and retention policies [5] is of great benefit in this area. Life-cycle costing principles were first developed in the US national defence industry, where traditionally budgeting for, say, a tank only took account of original manufacturing costs, not running, repair, maintenance or crew training costs. Life-cycle costing says that you look at the entire picture, the whole life of the item, then you know its real costs. Though the leap from defence procurement to library services is not one willingly made by many librarians, the truths of life-cycle costing are blindingly obvious. Each item of information in a storage systems has an initial cost, a time-dependent cost and a usage cost. It is a fact that librarians steal books. Probably a great many books. Estimates suggest that one in three thefts is an inside job. In 1973 a Library of Congress employee stole 2000 books worth $25,000. in 1982 an LA public library employee had 10,000 books and was lucky to have anything left of this floor. One of the UK's most famous recent thieves was Norma Hague, a librarian with a penchant for fashion plates who employed razor and cheesewire in the British Library, the V&A, and at Liverpool and Birmingham public libraries. Hundreds of plates were taken, estimated damage £50,000. Hague got two years with 16 months suspended. A famous non-librarian case was the Ethiopian Seymour McLean, who 'liberated' books he considered holy from the British Library, SOAS and other institutions. He got nine months, with the court rejecting his argument that he had stolen the items to return them to Ethiopia.