Bone marrow cells from two pediatric patients completing therapy for acute lymphoblastic leukemia were studied using in situ hybridization with an alpha-satellite DNA probe specific for chromosome 17. Morphologic analysis of the end-therapy specimens from each patient had shown small numbers (7.5%, 8.5%) of cells that were suspicious for residual or recurrent disease. These cells could not be morphologically or immunophenotypically distinguished with certainty from immature lymphoid cells (hematogones), which may be present normally, sometimes in increased numbers, in the bone marrow specimens of children. In situ hybridization with a probe to chromosome 17 was used because the leukemic cells from each patient had originally been shown to have an extra copy of this chromosome. In one patient, in situ studies showed a population of cells (106 of 1,000 cells) with three hybridization signals indicating trisomy 17, and thus residual/recurrent leukemia. In the other patient trisomy 17 could not be detected. Additional hybridizations to previously stained bone marrow aspirate smears permitted a direct correlation of the cytogenetic findings with the suspicious cells on a cell-to-cell basis. The questionable cells were identified, photographed, and then re-examined after hybridization. In one patient, 13 of 18 (72%) of the suspicious cells were found to have trisomy 17, whereas in the other patient 0 of 24 (0%) demonstrated an extra copy of this chromosome. These cases illustrate a clinical application of interphase cytogenetic analysis and demonstrate how this technology can be used for direct correlation of cytogenetic findings with cell morphology. This technique should prove useful for the detection of minimal residual disease and for lineage studies in leukemia and myelodysplasia.