Clonality assays may provide an additional and helpful approach to understanding the problem of the single or multistep pathogenesis of hematologic malignancies. In the absence of specific cytogenetic and/or molecular abnormalities, the study of X chromosome inactivation pattern in female patients represents a useful indirect approach for determination of the clonal or polyclonal nature of a cell population of unknown origin or the persistence of clonal hemopoiesis in leukemia after treatment. An intriguing observation was the documentation of an apparent clonal hematopoiesis in a quite high proportion of AML patients during morphological remission. These data were interpreted as evidence of a multi-step origin of AML, in which a first 'hit' was responsible for the growth advantage of a cytogenetically silent and morphologically normal preleukemic clone, whereas a second 'hit' would give rise to the overt leukemic phenotype. More recent surveys on the frequency of clonal remission, however, arrived at discordant results and could detect the occurrence of true clonal remission in a much lower percentage of cases than previously reported. These apparent discrepancies, which in part may be explained by differences in the series studied and in the approach used to detect clonality, for the moment leave open the fascinating issue of the single or multistep origin in many human leukemias.