Host shift and natural long-distance dispersal to an oceanic island of a host-specific parasite.

Parasite dispersal and host-switching may be better understood by knowing when they occurred. We estimated when the ancestor of a parasite of great reed warblers () dispersed to the Seychelles and began infecting the endemic Seychelles warbler (). We used mitochondrial genomes and published molecular divergence rates to estimate the date of divergence between mitochondrial haplotypes of the parasite (lineage GRW01) in the great reed warbler and the Seychelles warbler. We also constructed a time-calibrated phylogeny of the hosts and their relatives to determine when the ancestor of the Seychelles warbler dispersed to the Seychelles. The two GRW01 lineages diverged 20-451 kya, long after the ancestor of the Seychelles warbler colonized the Seychelles 1.76-4.36 Mya. GRW01 rarely infects other species despite apparent opportunity. Humans were likely not involved in the dispersal of this parasite because humans settled the Seychelles long after the parasite diverged from its mainland relative. Furthermore, introduced birds are unlikely hosts of GRW01. Instead, the ancestor of GRW01 may have dispersed to the Seychelles with an errant migrating great reed warbler. Our results indicate that even specialized parasites can naturally disperse long distances to become emerging infectious diseases.