This paper provides a method for the evaluation of the relative import
ance of different parts of extensive geographical areas for birds. It
uses data from the 'New Atlas of Breeding Birds: 1988-1991' at the 2-k
m and at the 10-km square scale in a case study of the uplands of nort
hern England. Six important bird areas are identified, each consisting
of a series of 2-km squares holding at least 1.5 times the mean regio
nal species-richness and importance index of upland birds. Their relat
ionships to existing Sites of Special Scientific Interest, Important B
ird Areas and proposed Special Protection Areas are discussed.