Criteria for the assessment of clonality by Southern blotting are well established but this is not the case for polymerase chain reaction (PCR)-based assays. Our studies, and infrequent reports in the literature, indicate that B-cell clonality may be erroneously inferred if only small numbers of polyclonal B-cells are present in test samples. In order to establish criteria to minimise the false positive assignment of B-cell clonality, DNA was analysed in a semi-nested PCR to detect rearrangement of the immunoglobulin heavy chain gene using a range (1 microgram-0.1 ng) of target DNA amounts from four tonsils and five lymph nodes showing reactive follicular hyperplasia, and from six B-cell lymphomas. A discrete, narrow band of PCR product of constant size was detected throughout the range of target DNA amounts in most lymphomas indicating the presence of a monoclonal B-cell population. In contrast, from the non-malignant tonsils and lymph nodes, larger target amounts generated a broad band of PCR products indicating populations of polyclonal B-cells, but smaller target amounts generated discrete, narrow PCR product bands of inconstant size indicating oligo- or monoclonal B-cell populations. Results of this study demonstrate that a range of DNA target amounts should be tested when the proportion of B-cells in a sample is unknown, thus preventing the analysis of insufficient target DNA which may lead to the false assignment of clonality.