Naive T lymphocytes specific for a given primary antigen occur in low frequencies and require the relevant antigen to be presented by specialist antigen presenting cells (APC), i.e., dendritic cells (DC). For these reasons, the in vitro induction of primary T lymphocyte responses remains a significant technical challenge. We have attempted to improve current strategies for generating in vitro responses by optimising (i) isolation and concomitant activation of DC from peripheral blood, (ii) uptake, processing and presentation of antigen by DC and (iii) antigen driven T lymphocyte proliferation. We established that RPMI 1640 media supplemented with 10% autologous serum resulted in the best yield of CMRF-44+, CD14-, CD19- DC after enrichment over a Nycodenz gradient. Optimal presentation of whole protein and peptide antigen was achieved by addition after the purification of the APC, i.e., at the start of the T lymphocyte proliferation assay. RPMI 1640 supplemented with 10% autologous serum or plasma supported the best antigen driven specific T lymphocyte responses. Using these optimised conditions, we compared the efficacy of PBMC and purified blood DC for priming T lymphocyte responses to the chronic myeloid leukaemia (CML) specific bcr-abl (b3a2) peptide. Peptide specific T lymphocyte responses were generated with both purified DC and whole PBMC, suggesting that T lymphocyte precursor frequency was the limiting factor in these experiments. These results will aid in the generation of human T lymphocyte lines to primary antigens, for in vitro and therapeutic applications.