When Edward Jenner introduced vaccination against smallpox in 1796, he was condemned for trying it out first on 23 experimental subjects, including an eight-year-old boy. The subjects were inocculated with the smallpox virus after vaccination to see whether immunity had been conferred. To opponents, this seemed an unforgivable risk of human life. Yet by 1806, after 10 years of vaccinations, London was able to record for the first time a full weak without a death from smallpox, and the disease has been suppressed since then wherever vaccination is practiced.
A similar weighing of short-term risk against long-term benefit is necessary whenever the storage environment for books and records is considered and the humidity level is specified. The risk involved in storage at low relative humidity (below, say, 40%,) is that of damage to already fragile materials if they are handled; the benefit is that of a significantly longer lifetime for all materials. Conversely, the risks involved in storage at high RH (over 50%. or so) are small, if mold and insect infestation can be controlled; but a shorter lifetime for these materials is a certainty.
In order to make an informed choice between alternatives, it is necessary to quantify the effects: How much of a risk is involved in handling dry materials? How much can the lifetime of materials be extended by consistent storage at low humidities? Where research is lacking, one has to rely, as always, on the guesses of experts and prevailing practice in related fields, although neither guide is much help in quantifying effects systematically. (Personal experience and tradition, which are often useful in other cases, are not reliable guides in this matter, because the effects that concern us are so subtle and so slow to reveal themselves that they can be detected only by controlled observation and measurement, usually with accelerated aging.)
However, one should not needlessly consult expert opinion or prevailing practice if controlled studies have been done, as they have for the effect of low RH on longevity. Finding, evaluating and summarizing the research is a job for an expert, especially if this review is to include a formula or formulas that would permit quantifying the effect on longevity of environmental conditions. It may be a difficult job, but not an impossible one, because so much is known already. One estimate made on the basis of research is that the longevity of paper is increased 50% by storage at 30% rather than 45% RH.
That figure for a 50% increase in longevity was based on results from accelerated aging, and reflects only the difference in RH. In real life, this increase would be much greater because of the way high RH speeds up the damage caused by light and air pollutants, factors not at work in the aging ovens. Oxidation (and therefore also hydrolysis, because the two processes are interdependent) is speeded in both natural and accelerated aging by higher RH.
Another effect of low humidity, one that runs counter to folk wisdom, is the prevention of brittleness in paper, leather and adhesives. Brittleness is caused by three mechanisms of aging: crosslinking, depolymerization and crystallization. All of these mechanisms are accelerated by higher relative humidity. The moisture content of paper does fall as it ages and as it is subjected to fluctuations in RH, so when old paper looks dry, it most likely is dry, but if this is seen as a problem, the solution should not be to store it in a damp environment.
One should not forget that several important factors, including the building itself and the climate in the given locality, set limits on the achievable range of RH. In dry, cold climates it may be impossible to push the RH above 20% on certain days, or even if possible it may be inadvisable because of the resulting condensation on poorly insulated windows and walls. In other climates, it may be impossible to keep the RH below 80% during the rainy season. It is likely that only a tiny percentage of all cultural institutions are able to hold the RH at a level they see as ideal, but even if so, the discussion of ideal RH for paper-based materials is not a purely academic one, because one still has to understand what is happening to the collection, and be prepared to upgrade if the opportunity presents itself.
Here is a collection of recommended storage RHs, with their sources, showing a lack of consensus that may reflect changing norms.