Barcode of Life Initiative: The Need
The classification and identification of living organisms is perhaps the oldest and most universal of the human sciences. Taxonomy is, and probably always has been, an element of human culture, reflecting both pragmatic considerations (e.g., the identification of edible, useful, or dangerous species) and the deeper human penchant for ordering natural diversity. The particular version of taxonomy practiced by modern science, with its system of binomial naming and hierarchical classification, has been under development since the mid-18th century.
Over the past 250 years, more than 1.7 million species of animals, plants, and other organisms have been described. While this is certainly an impressive accomplishment, it nevertheless represents only a small minority of the estimated 10-100 million species of eukaryotes alive today. Extrapolation of this past rate of discovery suggests that it will take another 1,500 to 15,000 years to complete the global inventory of life through conventional approaches. A second major taxonomic challenge arises from the need to recall the names and diagnostic morphological features of species once they have been formally described. In practice, learning the morphological nuances that separate closely allied species assemblages is so complex that few professional biologists, even those who have devoted their careers to taxonomy, can critically discriminate more than 1,000 species. The consequence is that to simply maintain a capacity for the identification of global diversity, we would require – in perpetuity – at least 10,000 taxonomists, all working on a different set of species (with the assumption that they all can be identified morphologically), with 100% replacement when these experts retire, and with easily accessible and employable morphological keys for all species lest these specialists do nothing but routine identifications. Even with only a minority of species described, the problem of obtaining accurate identifications is already acute for most biologists, educators, environmental regulators, and others, with a delay typically measured in months to acquire identities for even a short list of specimens – assuming that a qualified expert can be found.
All of this means that a rapid, accurate, automatable, and globally accessible procedure for species delimitation and identification is badly needed now, and will become even more necessary in the future. With the advent of efficient DNA amplification and sequencing methods, combined with advances in computing and information technology, we believe the time has come to implement a DNA-based system of species identification that is compatible with (and indeed, will both build upon and facilitate) the taxonomic infrastructure that has been assembled over the past 250 years.