Unfortunately, since very few collections have been able to make the primary data documenting their holdings available in a searchable digital format, straightforward, easy access to this massive information resource is currently not possible. Lack of digital accessibility to this critical information may come as a surprise since it is clear that many countries, particularly ones with large gross domestic products and signifi cant research budgets, could easily support the development of such digital access, and this access would prove to be very benefi cial for their citizens and the world as a whole. In fact, without access to such a searchable digital catalog of the holdings of any particular institution, it is very diffi cult to assess and predict the taxonomic, temporal, and geographic scope of that institute’s holdings. This is due to the fact that the holdings of few natural history collections have been accumulated based on any long-term strategic planning process (Lane, 1996). Additions to any natural history collection holdings at any particular time have been dependent on the taxonomic and geographic interests of particular staff members, and the availability of support for the staff’s collecting and curation activities. They are a refl ection of ad hoc science, which cannot be judged as effi cient and hypothesis-driven. Thus, the taxonomic, geographic, and temporal distribution of the holdings of any collection refl ects changes in staff and staff interests, and funding support for collecting activities throughout the history of that collection. As a result of this erratic way in which any particular collection has developed, it

is impossible, without having someone physically study the collection, to assess and predict the taxonomic, temporal, and geographic distribution of its holdings, or its scientifi c merit even, and how to improve it. Although it is usually possible to identify individual natural history collections with signifi cant holdings covering particular taxonomic groups, geographic areas, or specifi c time periods, these are likely to represent only a small proportion of the records that would be available if one could access the holdings of all natural history collections simultaneously. In fact, unless their holdings have been databased, very few institutions themselves are even knowledgeable about the distributions of their holdings. Peterson and Navarro-Sigüenza (2003) identify the documentation of a collection’s holdings as one of the most signifi cant benefi ts of databasing a collection. However, many other uses can be named, but many of them are just developing with the development in other disciplines (Graham et al., 2004) and are not even foreseeable.