The possibility to perform cross-species microsatellite amplification in birds was surveyed by analysing sets of primers developed from the swallow and the pied flycatcher genomes on a panel of 48 different bird species. In total, 162 cases (species/marker combinations) of heterologous amplification were recorded. Ten amplification products were sequenced and all were found to be true homologues of the original loci. There was a significant and negative relationship between microsatellite performance and evolutionary distance between the original species and the tested species. As a rough indicator of expected cross-species microsatellite performance we estimate that 50% of markers will reveal polymorphism in a species with a DNA-DNA hybridization delta T(m)H value of 5 separating it from the original species. This corresponds to a divergence time of = 11 million years before present for passerine birds. The established relationship between performance and evolutionary distance agrees very well with data obtained from some mammalian species. The proportion of polymorphic loci among those markers that amplified decreased with increasing genetic distance, suggesting that few long repeats are preserved during evolution. One of the swallow markers, HrU2, amplified a specific product in all species analyzed and will thus allow access of nuclear sequence data over a broad range of species. The only predictor of cross-species performance was the amount of non-specific amplification seen in the original species. An analysis of 10 species from within the family Hirundinidae with the swallow primers consistently revealed extensive polymorphism with average probabilities of identical genotypes ranging from 6 x 10(-4) to 6 x 10(-7). There were distinct allele frequency differences between the Hirundinidae species and we envisage that microsatellite cross-species amplification will be a useful tool in phylogeny construction and in species identification.