The mechanism of aberrant genetic recombinatorial events in neoplasia is vaguely understood. One hypothesis is that aberrant DNA methylation may in some way predispose genomic regions to recombination. Using the t(9;22) of chronic myeloid leukemia (CML) and the t(15;17) of acute promyelocytic leukemia (APL) as models, our laboratory and others have gathered data supporting this hypothesis. These data are reviewed.