Towards a digital African Flora

Modern approaches to Flora writing are increasingly being built around database technology. Databases allow for great flexibility in the range of questions that can be asked of the Flora, and consequently greatly enhance the utility of the Flora in the core roles of identification of species and nomenclatural queries. There are, however, few examples where existing floristic works have been converted into database format, so that much of our legacy of floristic texts still rests in the form of the printed page. In this paper we address the problem of the digitisation of legacy texts (see www. kew. org/efloras/). We describe a method that we used to create an online database of the Flora Zambesiaca for identification and nomenclatural queries. We assert that the method we used can be generalised to cope with most standard taxonomic formats. We conclude that on a costbenefit basis Flora digitisation compares very favourably to other digitisation approaches, such as specimen databasing and interactive key construction, and could be used with good effect to direct digitisation in these other areas.