There exist several examples of mobile group I introns. These introns appear to use a straightforward mechanism to achieve highly site-specific and efficient insertion into homologous intronless genes. Because the only intron-specific function required by the prevailing model for the mechanism of intron mobility is the introduction of a site-specific double-stranded break in the intronless recipient DNA molecule, we reasoned that it should in principle be possible to construct artificially mobile DNA sequences. We have constructed an artificial mobile element from the gene for the restriction enzyme EcoRI that is capable of site-specific insertion at rates near those of authentic mobile introns. The generality of the mobility mechanism may enable high-efficiency targeted gene replacements or disruptions in a variety of organisms.