The immunological detection of minimal residual disease in B-lineage acute lymphoblastic leukaemia (ALL) has been hampered by the fact that the leukaemic cells represent the malignant counterparts of normal haemopoietic precursors expressing terminal deoxynucleotidyl transferase (TdT), CD10 and CD19. We have used quantitative double-labelling flow cytometry with standard fluorescent beads to convert the mean fluorescence to the number of antigen molecules per cell. The number of TdT, CD10 and CD19 molecules per cell was determined in normal B-cell precursors from 22 healthy donors and eight regenerating marrows from patients with various malignancies and in 20 cases of B-lineage ALL. In normal bone marrow we characterized two different B-cell populations: TdT+/CD10+/CD19+ and TdT-/CD10+/CD19+. We demonstrated a major difference in the level of expression of TdT, CD10 and CD19 between normal bone marrow and B-lineage ALL blasts. Normal TdT+ precursors have significantly higher number of TdT (> 100 x 10(3)) and lower number of CD10 (< 50 x 10(3)) and CD19 (< 10 x 10(3)) molecules per cell than B-lineage ALL blasts (< 100, > 50, > 10 x 10(3) molecules per cell respectively); these differences were statistically highly significant. Furthermore, regenerating marrows had a significantly higher percentage of B-cell precursors than healthy donors. This increase was at the expense of the TdT-/CD10+/CD19+ population which, in the context of B-lineage ALL, could be wrongly interpreted as evidence of relapse if TdT is not included in the analysis. Therefore the quantitative analysis of TdT combined with CD10 and CD19 may allow a clear distinction between normal precursors and minimal residual leukaemia in B-lineage ALL and avoid the pitfall of misinterpreting regenerating B-cells as evidence of relapse.