Diagnosable versus distinct: evaluating species limits in birds

I review new species of birds described from 1993 to 2002, comparing the methods used to diagnose (identify) new bird species with those used for other classes of vertebrates. Birds are described and subsequently identified using a far more restrictive set of criteria than other organisms, with field diagnosability being the ultimate criterion of distinctiveness. This has led to a large number of evolutionarily distinct bird taxa that are not recognized as distinct taxonomic units. This disparity has important consequences for ornithology and for other disciplines that use birds as biodiversity measures.