Identification of rearrangements in the immunoglobulin heavy chain (IgH), T cell receptor (TCR) and other genes assists in the classification of hemopoietic disorders. This study reviews a 5 year experience of genotyping as an assessment of clonality in a central referral laboratory. Patients were referred for assessment if abnormal hemopoietic populations were identified and where standard morphology, cytochemistry or immunophenotyping techniques were equivocal and unable to establish the diagnosis. Analysis of the antigen receptor genes (IgH gene and TCR gamma locus) was performed in 230 patients by either Southern blotting or the polymerase chain reaction (PCR). Clonal rearrangements of either loci could be demonstrated in 91/230 patients: 56/161 patients analyzed by Southern blotting and 48/125 analyzed by polymerase chain reaction. A subgroup of patients (n = 88) was analyzed for rearrangement of the immunoglobulin heavy chain gene by both techniques. Discordant results were observed in 18 of these patients (20%). Analysis of the TCR gamma locus in a separate group demonstrated discordant results in eight of 40 patients examined (20%). Clinical outcome could be available in 61 patients (median time: 42 months, range 1-68 months): for those in whom a rearrangement was detected approximately 80% went on to develop a lymphoproliferative disorder although this diagnosis was not able to be made at the time the sample was taken. In a subset of patients (n = 14) who presented with lymphocytosis after bone marrow transplantation (seven) or solid organ transplant (seven), clonal lymphoid populations were demonstrated in approximately 50% of cases. The majority of these cases demonstrated the presence of Epstein-Barr virus (EBV) by PCR. The clonality of EBV infection was assessed by Southern blotting and the significance of these findings are discussed. This study explores the differences between Southern blotting and PCR as applied to the study of antigen receptor rearrangement studies. We conclude that detecting a clonal population is valuable in patients where standard diagnostic techniques are equivocal. However, the inability to detect a clonal population should be treated with caution and interpreted in light of other investigations.