Can species data only be appropriately used to conserve biodiversity?

Despite widely acknowledged handicaps of the species approach to identifyin g priority conservation areas, many workers continue to use these flawed te chniques as the backbone of their analyses. Species-based approaches addres s only a small part of biological diversity by ignoring different levels of organisation as well as the functional linkages among these levels. These data are often biased and incomplete and are often used in preference to da ta dealing with higher biological levels of organisation though the latter may be readily available. Within the framework of Noss's [(1990) Conservati on Biology 4: 355-364] hierarchical definition of biodiversity (and Scott e tal. [(1993) Wildlife Monographs 123: 1-31] gap analysis), we propose a top -down model dealing with broad organisational levels first, and finer-scale species distributions last. Note that we do not discard the latter approac h, but merely argue for its use at a stage when, in our opinion, it adds mo st to the value of the prioritisation exercise. The model is flexible so th at additional information, particularly those related to threats to biologi cal diversity, can be added when they are available.