THERE are over three million books and manuscripts in the libraries of Trinity College. A large portion of these date back to the early centuries and have suffered structural damage either from careless handling, pollution, strong light or simply through the ageing process. All of which poses a threat to their remaining shelf life.

In 1974, a conservation laboratory was set up to halt any further deterioration. The department is run by Anthony Cams, a former technical advisor to the Bibloteca Nazionale Centrale in Florence - a conservation centre set up by the UK Art and Archive Rescue Fund in 1966. There he worked with a staff of ninety but budgetary constraints mean he has to be content with a full time staff of three at Trinity.

Raymond Jordan (studied in Bolton Street) and Matthew Hattan (apprenticed at the Victoria & Albert museum, London) are the senior technicians; John Gillis was trained as a bookbinder in Bolton Street and, as the lab's junior technician, is now being trained in book conservation. With them for a year is Anna Stenstrom, a student intern from the School of Library Service, Columbia University, New York.

The basic problem faced by the conservation laboratory is stabilising the degraded material in .both physical and chemical terms. Leather manuscript binding contains traces of iron and copper which act as catalysts in converting sulphur dioxide - found in polluted atmospheres - to sulphur trioxide, thus forming sulphuric acid which eats away at the leather. Moreover, the material has, over the years, developed a dangerously high acid content which threatens its permanence.

The lab's first task is to neutralise the paper by giving it a pH value of between seven and a half and eight. Uncontrolled neutralising is avoided however as there is an additional risk of oxidising the cellulose material if the pH rises above nine.

(While it is universally accepted that any material with a pH less than 5.5 is in need of urgent treatment, much of the library's collection has fallen well below this level). Alkalinity can be induced either through aqueous or non-aqueous methods, the latter usually being used when the manuscript ink is sensitive to water. With the aqueous method, the material to be treated is sandwiched between layers of egg crate light diffuser, acrylic and polyester web to prevent any of the damaged material, should it tear from the main text, from being lost.

It is then washed in a buffering solution containing calcium hydroxide with calcium bicarbonate being added to give it a calcium reserve for alkalinity. Many of the early manuscripts have been reinforced with rosin-alum sizing which tends to inhibit the buffering solution, so these are usually pre-wetted with ethanol alcohol and water or, alternatively, iso-propanol to allow penetration. When the lab cannot afford to take the manuscript apart, non-aqueous methods are used. Previous treatments by labs abroad involved the use of magnesium methoxide which although it led to a good alkaline reserve was very sensitive to water.

Trinity's lab uses methylmagnesium carbonate which has been found to be much more stable to changes in the environment and has a wider tolerance to different application techniques. It is applied with a brush. After treatment, the material is then reinforced with Hydroxypropyl cellulose (Klucel G') for added strength. Because the manuscripts in the library's Long Room are bound in leather, they have been particularly prone to red rot and have become very reactive with water.

A humid cloth is therefore used to remove the surface soil from the binding while the bookshelves are vacuum cleaned and lightly dusted. The leather also needs to be stabilised or consolidated to stop it powdering off. Since water based consolidants would ruin the leather, the lab had to come up with an alternative which would neither darken nor discolour the degraded leather. They decided to dissolve the Klucel 'G' in a 1% solution of Anhydrous Ethanol and found it met with their requirements. This discovery is now used by many other conservation labs abroad. The problem of protecting leather is still receiving attention both here and abroad. It may be solved by applying reagents to deprive the iron contained in the leather of its power to catalyse the formation of sulphuric acid.

The British Leather Manufactifring Research Association is currently recommending the application of Aluminium Triformate which is effectively a mineral retannage of leather. It can only be applied to very new vegetable tanned leather as degraded leather would immediately turn black. At present a chromium and sumac mixture is used as a leather tannage; and is much more stable than earlier tannages such as oak.

Text is sacrosanct and, if repairs are needed, only the damaged area is -touched. Special thermoplastic repair tissue, such as good quality lens tissue, impregnated with a mixture of polyacrylate resins is used. A silicon coated release paper is used as an intermediatary layer as the tissue is set with an iron heated to about 800C. The important aspect of this process is that -unlike earlier methods - it can be easily reversed using solvents such as acetone or sometimes simply water.

Before the war, inter-leaving was carried out with nitro cellulose but this turned out to be disastrous with the paper turning as black as carbon. Now inert polyester films are used.

The text may need to be rebound. For this, textiles such as Ulster Linen are used. They are incredibly strong, difficult to tear and, more importantly, have a neutral pH. After the cleaning, stabilising and rebinding, the finished text is placed in a special 'bookshoe' slip case made from very strong neutral (pH of 8) board, usually 'Permalife'. This helps to keep the book intact while still allowing the spine to be seen so that the aesthetic and architectural value of the Long Room is retained - an important feature of conservation policy.

All paper and board used in the process must be neutral and has to be imported since there isn't any mill in Ireland producing neutral paper. According to Anthony Cams, this reduces the life of any maps or books made in this country.

The lab has also worked on a number of vellum manuscripts, the most recent being the Life of St. A/bans, a 13th century manuscript, and the Winchcombe Psalter, dating back to the 12th century and recently displayed at the Romanesque exhibition in London.

Great care must be taken when working with vellum. For a start it is very sensitive to changes in the relative humidity and can swell or shrink rapidly and cause the pigment to detach. Hard pressure can make the skin transparent and when damp the writing may print off or mildew could form.

When vellum was originally made it was first damped and then stretched on a frame and allowed to dry under tension. If it becomes damp again however it tends to revert back to the original shape of the animal. Thus many of the ancient works have become terribly cockled. However. They can easily be reflattened by redamping, stretching and drying, without endangering the text.

Degreased fish skin or rather its swim bladder - a transparent membrane - is used for repairs to vellum. Any resewing will be carried out with linen thread and will keep to the original sewing positions ensuring the leaves remain untouched by the fresh needle.

The rebound manuscripts are placed in special wedge-shaped boxes, normally made from oak, teak ,or pine, which keep a light restraining pressure on the leaves. Because of the amount of time involved, the repairing of a manuscript can cost anything from £4,000 and upwards.

Although a lot of work appears to be carried out on the texts, Anthony Cams prefers to describe the whole process as one of "minimum intervention" in that only very necessary work is carried out. He shuns the idea of restoration and says: "we don't perfect and imperfect the object, nor do we try to hide what we have done. All of our work will be recorded in a log for any particular item".

The department is also concerned with preventing any further deterioration of the historic materials and has taken steps to control the surrounding conditions.

Strong light entering the Long Room has been one of the main sources of damage to the leather bound texts and the windows have now been shielded with UV filters. The display cases for exhibiting the manuscripts have also been screened with UV filters and several protective layers of glass. Inside the closed atmosphere of the case, relative humidity is carefully con- trolled by using silica gel specially pre- conditioned to act as a moisture buffer rather than as a pure drying agent. This keeps the Equilibrium Moisture Content (EMC) constant within the display case.

Relative humidity has always been a problem and steps have been taken to ensure it's carefully controlled in the manuscript strong room. In fact a study by the British Museum found that every 1,000 tons of books absorbed over 20,000 lbs of water when the relative humidity was increased from 57% to 63% at 6O0F(160C). Of course the ideal conditions for storing paper are complete darkness and zero humidity but it would make the paper too brittle for reading purposes.