The use of sounds in avian systematics and the importance of bird sound archives

SUMMARY The steady increase in the global total of bird species is in part due to the discovery ofdistinct vocalisations which reveal hitherto unrecognised information, either the presenceof entirely new species or the level of differentiation in taxa previously treated asconspecific. There are many examples of both types of discovery (and some where taxapreviously given species status prove conspecific). The importance of vocal analysis inestablishing the relationships between taxa is also demonstrated by numerous examples.Avian sound archives are clearly crucial to the advancement of such studies, but manymore recordings are needed, their quality must be high, the circumstances of recordingsneed to be documented, the identity of vocalising birds needs to be firm, and a fullersystem of cooperation between sound archives is required. Introduction The number of bird species in the world was estimated to be c.8,600 by Mayr (1946),c.9,000 by Bock & Farrand (1980), and c.9,700 by Sibley & Monroe (1990). Thisincrease widely exceeds the number of newly described species in that time period.Part of this increase can be attributed to the growing knowledge of bird vocalisationsin combination with the current trend to recognise allopatric taxa with distinctivesongs as species rather than as subspecies (Sibley & Monroe 1990, Parker 1991,Price 1996, Peterson 1998).Although sounds have mainly been of use in the ranking of closely relatedallopatric taxa, vocalisations have also been used to infer relationships, both withinand between genera, and have been crucial in the discovery of several new species.We here review the use of vocalisations and other acoustic signals in systematics;see also Payne (1986) for a thorough review and Morel & Chappuis (1992) for areview of West African taxa. We also discuss the importance of sound archives andsuggest how they may be made more useful to future researchers.